Back to EveryPatent.com
United States Patent |
6,036,827
|
Andrews
,   et al.
|
March 14, 2000
|
Electrolyzer
Abstract
The present invention provides a method and apparatus for chemically
heating a catalyst bed by feeding hydrogen to the catalyst. The invention
also provides a method and apparatus for thermally conditioning a catalyst
in order to enhance the conversion of unacceptable emissions (emanating
from an internal combustion engine) into water and other acceptable
emissions. In one aspect of the invention, hydrogen is supplied from an
electrolyzer or other hydrogen source and injected into the monolith of a
catalytic converter to more rapidly bring the catalyst to a light-off
temperature.
Inventors:
|
Andrews; Craig C. (College Station, TX);
Murphy; Oliver J. (Bryan, TX)
|
Assignee:
|
Lynntech, Inc. (College Station, TX)
|
Appl. No.:
|
884112 |
Filed:
|
June 27, 1997 |
Current U.S. Class: |
204/252; 204/253; 204/257; 204/258; 204/266 |
Intern'l Class: |
C25B 009/00 |
Field of Search: |
204/252,263-266,283,282,253,257,237
|
References Cited
U.S. Patent Documents
3311097 | Mar., 1967 | Mittelstaedt.
| |
3420052 | Jan., 1969 | Miller.
| |
3719457 | Mar., 1973 | Nagamatsu.
| |
3729936 | May., 1973 | De Palma et al.
| |
3761229 | Sep., 1973 | Schwartz.
| |
3779014 | Dec., 1973 | Nohira et al.
| |
3782115 | Jan., 1974 | Johnson.
| |
4332219 | Jun., 1982 | Gonzalez.
| |
4368696 | Jan., 1983 | Reinhardt.
| |
4499864 | Feb., 1985 | Lovercheck et al.
| |
4615775 | Oct., 1986 | de Nora et al. | 204/252.
|
4685430 | Aug., 1987 | Ap.
| |
4763610 | Aug., 1988 | Thomas.
| |
4865818 | Sep., 1989 | Merry et al.
| |
4939902 | Jul., 1990 | Retallick.
| |
4985210 | Jan., 1991 | Minami.
| |
5130109 | Jul., 1992 | Wan.
| |
5155995 | Oct., 1992 | Kinnear et al.
| |
5163290 | Nov., 1992 | Kinnear.
| |
5184462 | Feb., 1993 | Schatz.
| |
5207734 | May., 1993 | Day et al.
| |
5259190 | Nov., 1993 | Bagley et al.
| |
5263318 | Nov., 1993 | Oota et al.
| |
5272871 | Dec., 1993 | Oshima.
| |
5419121 | May., 1995 | Sung et al.
| |
5813222 | Sep., 1998 | Appleby.
| |
Foreign Patent Documents |
73 14944 | Dec., 1974 | FR.
| |
2246218 | Mar., 1974 | DE.
| |
2 387 357 | Nov., 1978 | DE.
| |
4208609A1 | Mar., 1992 | DE.
| |
4208609 A1 | Mar., 1992 | DE.
| |
4103668 A1 | Aug., 1992 | DE.
| |
41 03 668 A1 | Aug., 1992 | DE.
| |
44 20 715 A1 | Jan., 1995 | DE.
| |
62-045922 | Feb., 1987 | JP.
| |
2-086915 | Mar., 1990 | JP.
| |
4214183A1 | Apr., 1992 | JP.
| |
4-318214 | Nov., 1992 | JP.
| |
4318214 | Nov., 1992 | JP.
| |
6-033748 | Feb., 1994 | JP.
| |
6033748 | Feb., 1994 | JP.
| |
93/06850 | Jul., 1993 | WO.
| |
WO96/11330 | Apr., 1996 | WO.
| |
Other References
Youji Kanada, Masaharu Hayasi, Motonobu Akaki, and Shunzou Tsuchikawa,
"Hydrogen Added After-Burner System," 1996, Society of Automotive
Engineers, Inc., pp. 89-94 (No Date).
|
Primary Examiner: Valentine; Donald R.
Attorney, Agent or Firm: Streets & Associates, Streets; Jeffrey L.
Claims
What is claimed is:
1. An electrolyzer comprising:
an anode compartment comprising a plurality of rigid components including
an anode flowfield and a porous anodic electrocatalyst substrate molded
into a cell frame;
a cathode compartment comprising a resilient member;
a flexible membrane disposed between the anode and cathode compartments.
2. An electrolyzer comprising:
an anode compartment comprising a plurality of rigid components including
an anode flowfield and an anode product outlet in communication with an
upper portion of the anode flowfield;
cathode compartment comprising a resilient member;
a flexible membrane disposed between the anode and cathode compartments;
and
an upwardly sloping passage in communication between the anode product
outlet and a water reservoir.
3. An electrolyzer comprising:
an anode compartment comprising a plurality of components including an
anode flowfield and anode product outlet communicating with an upper
portion of the anode flowfield;
a cathode compartment;
a flexible membrane disposed between the anode and cathode compartments;
and
an upwardly sloping passage in communication between the anode product
outlet and a water reservoir; and
a bubble detector disposed adjacent to the upwardly sloping passage.
4. An electrolyzer comprising:
an anode compartment comprising a plurality of components including an
anode flowfield and anode product outlet communicating with an upper
portion of the anode flowfield;
a cathode compartment; and
a flexible membrane disposed between the anode and cathode compartments;
and
an upwardly sloping passage in communication with the anode product outlet.
5. The electrolyzer of claim 4, further comprising a bubble detector
disposed adjacent to the upwardly sloping passage.
6. An electrolyzer comprising:
an anode compartment comprising a plurality of components;
a cathode compartment comprising a cathode, a cathode flowfield and a
cathode product outlet communicating with an upper portion of the cathode
flowfield;
a flexible membrane disposed between the anode and cathode compartments;
and
an upwardly sloping passage in communication with the cathode product
outlet.
7. The electrolyzer of claim 6, further comprising a bubble detector
disposed adjacent to the upwardly sloping passage.
8. The electrolyzer of claim 6, further comprising:
a hydrogen storage vessel having an inlet in communication with the cathode
product outlet.
9. The electrolyzer of claim 8, wherein the hydrogen storage vessel has an
inlet that is elevated above the cathode product outlet.
10. The electrolyzer of claim 8, further comprising a check valve disposed
in fluid communication between the cathode product outlet and the hydrogen
storage vessel to prevent backflow from the hydrogen storage vessel to the
cathode.
11. The electrolyzer of claim 8, further comprising:
a water reservoir in fluid communication with one of the electrode
compartments; and
a water recycle line communicating water from the hydrogen storage vessel
to the water reservoir.
12. The electrolyzer of claim 8, further comprising:
a check valve disposed to prevent backflow from the hydrogen storage vessel
to the cathode.
13. The electrolyzer of claim 6, further comprising:
a hydrogen storage vessel having an inlet in communication with the
upwardly sloping passage.
14. An electrolyzer comprising:
a first electrode compartment comprising a rigid component;
a second electrode compartment comprising a resilient member;
a flexible membrane disposed between the first and second electrode
compartments; and
a water reservoir in fluid communication with one of the electrode
compartments.
15. The electrolyzer of claim 14, wherein the water reservoir is a
windshield wiper fluid reservoir.
16. The electrolyzer of claim 14, wherein the water reservoir communicates
with an anode through an anode flowfield.
Description
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to the field of catalysis for the
reduction of emissions from internal combustion engines. More
particularly, the present invention relates to a method and apparatus for
producing hydrogen and oxygen, storing hydrogen and delivering hydrogen
into a catalytic converter on a vehicle in order to heat the catalyst
through catalytic combination of the hydrogen with oxygen provided or from
the air or stored oxygen. This spontaneous catalytic combination releases
sufficient heat to rapidly bring the catalytic converter to operating
temperatures.
2. Background of the Related Art
The control and suppression of unwanted emissions created by the operation
of an internal combustion engine is a primary consideration for engine
designers and vehicle manufacturers because of nearly world-wide
governmental requirements regarding acceptable emission levels. Over
eighty percent (80%) of the unacceptable emissions or pollutants created
by internal combustion engines equipped with catalytic converters occur
during cold start operations. These pollutants are emitted for a period of
one to three minutes after cold engine starting, in large part because
that is the time period required for the catalyst to reach an efficient
operating temperature. Therefore, even though the engine exhaust is
flowing through the catalytic converter, until the exhaust heats the
catalytic converter to its operating range from engine start up, the
engine emissions are only slightly catalyzed during that time period.
In order to meet governmental emission standards for internal combustion
engine exhaust a catalytic converter is located in the exhaust stream of
the engine. The converter typically includes a canister holding a suitable
catalyst, such as a three-way catalytic converter (TWC) catalyst monolith,
that will oxygenate unburned, unacceptable components in the exhaust
stream including hydrocarbons (HC), their partially oxidized derivatives
such as aldehydes and carbon monoxide (CO), and at the same time reducing
nitrogen oxides (NO.sub.x), after almost stoichiometric fuel burn with
oxygen in the cylinders of the engine. The exhaust gas is passed through
the catalyst monolith, thereby completing the oxygenation of unburned HC
and CO, and the reduction of NO.sub.x in the exhaust to convert these
unacceptable emissions into acceptable emissions. Certain unacceptable
emissions in the exhaust stream, including unburned hydrocarbons and
carbon monoxide, require an oxidation reaction to destroy them so that
they end up as the corresponding oxides, e.g., water and carbon dioxide.
On the other hand, NO.sub.x requires a reduction reaction to develop
N.sub.2 and O.sub.2. In fact, the O.sub.2 product of this reduction
contributes to the oxidation of the HC and CO in the exhaust.
TWC catalysts are currently formulated and designed to be effective over a
specific operating range of both lean and rich fuel/air conditions and a
specific operating temperature range. These particular catalyst
compositions enable optimization of the conversion of HC, CO, and
NO.sub.x. This purification of the exhaust stream by the catalytic
converter is dependent on the temperature of the exhaust gas and the
catalytic converter works optimally at an elevated temperature, generally
at or above about 300.degree. C. Light-off temperature is generally
defined as the temperature at which fifty percent (50%) of the emissions
from the engine are being converted as they pass through the catalyst. The
time period between "cold start" and reaching the light off temperature is
generally referred to as the light-off time.
The conventional method of heating the catalytic converter is to heat the
catalyst by contact with high temperature exhaust gases from the engine.
This heating, in conjunction with the exothermic nature of the oxidation
reaction occurring at the catalyst, will bring the catalyst to light-off
temperature. However, until the light-off temperature is reached, the
exhaust gas passes through the catalyst relatively unchanged. In addition,
the composition of the engine exhaust changes as the engine heats from the
cold start temperature, and the catalyst monolith is typically designed to
work best with the composition of the exhaust stream produced at the
normal elevated engine operating temperature.
There have been several attempts to shorten or avoid the light-off time of
the catalytic converter. Current techniques employ one or more of the
following methods: electrical heating of the exhaust gases and/or of the
catalytic converter itself; thermal insulation; multi-chambered
configurations of the catalytic converter; and/or placing the catalytic
converter adjacent to the engine for heating. All of these methods have
drawbacks and limitations.
Placing the catalytic converter almost immediately adjacent to the engine
is not desirable because of the tendency to overheat the catalyst with
resulting accelerated degradation of the catalyst. Thermal insulation is
also not a desirable option because of the same problems, especially
during operation under maximum operating temperature ranges.
Electrical heating of catalytic converters ("EHC") has been a popular
proposed method of attempting to preheat the catalyst monoliths.
Limitations on the equipment and process, however, affect the utility of
this method. The primary limitation on electrical preheating is the
electrical energy required by the heater. The typical car battery is not a
practical power source to supply the electrical power because the
electrical load on the vehicle battery during the period required may
exceed the rated battery output. In any event, the load placed on a
typical 12 volt vehicle battery will shorten the lifetime of the battery.
Also, there is a measurable delay between the time the operator of the
vehicle places the ignition switch in the "on" position and the time the
heater brings the catalyst to light-off temperature.
Typically, in the interval between start up and light-off, the exhaust
stream is oxygen deficient. Because the catalyst requires oxygen to
complete the catalytic reaction, supplemental air must be blown over the
catalyst. Even when using a secondary air flow to overcome oxygen
deficiency, the secondary air flow must be closely controlled to avoid an
excess of oxygen, in which case the catalytic converter is less effective
in reducing NO.sub.x. However, it should be noted that NO.sub.x
contributes a very small portion of unacceptable emissions when an engine
is cold; most of the cold start emissions that must be dealt with comprise
HC, CO and the like.
An alternative to battery powered electrical heating has been to decrease
the strain on the power supply by supplying the power directly from an
alternator rather than directly from the vehicle battery. An alternator
powered, electrically heated catalyst ("APEHC") still requires a 5 to 10%
increase in battery capacity to cope with the EHC start-up scenario. Even
with the APEHC system, there is still a concern with respect to battery
capacity because electric heating is needed for an extended period of
time, i.e., more than 25-30 seconds. In addition, the maximum alternator
power output required in the APEHC system requires a complicated switching
mechanism and an altered alternator speed between 2,000 and 4,500 rpm
during the heating up time period, and the alternator must be oversized.
The multi-chamber configurations of catalytic converters generally conform
to one or two theories. In one multi-chamber configuration, a small
portion of catalyst known as a "starter catalyst" is positioned upstream
from the primary catalyst. This "starter catalyst" is generally closer to
the exhaust manifold. This location, in conjunction with a smaller thermal
mass associated with its smaller size and materials of construction,
causes the catalyst to heat much more quickly than a single catalyst. This
configuration, however, is generally unacceptable because the starter
catalyst in the exhaust stream creates a higher back pressure which
reduces the overall engine efficiency and robs the engine of power output.
Another method of providing multiple chambers in the exhaust flow includes
a first catalyst having low temperature characteristics used only during
cold start conditions, and, after the catalyst temperature rises to a
certain elevated level, the exhaust gas flow is switched to pass through
the conventional catalytic converter configuration. A variation of this
approach is to run all cold start emissions through a separate absorber
(such as a zeolite or a metal sieve-type substance) where unacceptable
emissions are captured and later released back into the exhaust stream.
This method, however, is impractical because of the complicated switching
mechanism used to divert flow to the absorber, the size and space
requirements of the absorber, and the impracticality of releasing the
unacceptable emissions from the absorber back into the exhaust stream.
Finally, one additional method for reducing cold start emissions runs the
engine excessively rich in the cold start condition and ignite the
resulting super-rich mixture to directly heat the catalyst. This approach
has proved wholly unreliable and has other serious drawbacks, including
reduced engine and catalyst life.
To date, there has not been a catalytic converter heating system which
gives almost instantaneous heating of the catalytic converter without the
inherent drawbacks stated above. Thus, there remains a need for an
improved catalytic converter system that reduces ineffective catalytic
action immediately after cold start-up of an engine. Such a system must be
simple and must not reduce the rated lifetime of the engine, the catalytic
converter, or the battery components of the vehicle.
SUMMARY OF THE INVENTION
The present invention provides an on-board system for the generation of
hydrogen, comprising an electrolyzer having an anode in communication with
a water reservoir, a cathode, and a proton exchange membrane disposed
between the anode and the cathode; a source of electrical current having a
positive terminal coupled to the anode and a negative terminal coupled to
the cathode; and a controller in electronic communication with the source
of electrical current. The system may also comprise a hydrogen storage
vessel having an inlet in communication with the cathode and an outlet in
communication with a catalytic converter. Preferably, the cathode
comprises a cathode flowfield and cathode product outlet communicating
with an upper portion of the cathode flowfield. In one embodiment, an
upwardly sloping passage may be incorporated to provide communication
between the cathode product outlet and the hydrogen storage vessel and a
bubble detector disposed adjacent the upwardly sloping passage to detect
the production of gases by the electrolyzer. It is also preferred that the
system include a check valve disposed in fluid communication between the
cathode and the hydrogen storage vessel to prevent backflow from the
hydrogen storage vessel to the cathode.
The water reservoir, such as a dedicated reservoir, a windshield wiper
fluid reservoir or a hydrogen storage vessel, may communicate with the
anode through an anode flowfield and/or through the proton exchange
membrane. The source of electrical current may comprise a battery or an
alternator. The hydrogen storage vessel preferably includes a pressure and
a water recycle line communicating water from the hydrogen storage vessel
to the water reservoir.
The invention also provides an automobile, comprising an internal
combustion engine having an exhaust system; a catalytic converter disposed
within the exhaust system; an electrolyzer having an anode in fluid
communication with a water reservoir, a cathode, and a proton exchange
membrane disposed between the anode and the cathode; a source of
electrical current having a positive terminal coupled to the anode and a
negative terminal coupled to the cathode; and a pressure vessel in
communication with the cathode. The automobile will preferably further
comprise a controller in electronic communication with the source of
electrical current, a source of oxygen gas in communication with the
catalytic converter and a flow control member disposed between the
pressure vessel and the catalytic converter for delivering a cathode
product to the catalytic converter. wherein the flow control member
provides pulsed flow of the cathode product. In one embodiment, the
automobile will include an ignition system in electronic communication
with the internal combustion engine and the flow control member.
The invention further provides a method of monitoring the performance of an
electrolyzer, comprising the step of detecting gas bubbles passing through
an electrolyzer outlet passage or channel.
The invention also provides an apparatus for pulsed delivery of hydrogen
from a pressurized hydrogen source to a vehicle exhaust system, comprising
a gas passageway communicating between the hydrogen source and the exhaust
system; first and second isolation valves disposed in the gas passageway,
the second valve being downstream from the first valve; and a region
between the first and second isolation valves having a defined volume.
Alternatively, the invention provides an apparatus for pulsed delivery of
hydrogen from a pressurized hydrogen source to a vehicle exhaust system,
comprising a turnstile valve disposed in a conduit between the hydrogen
source and the exhaust
An electrolyzer is also provided, the electrolyzer comprising a first
electrode compartment comprising substantially incompressible components;
a second electrode compartment comprising a compressible member; and a
flexible membrane disposed between the first and second electrode
compartments. This electrolyzer is particularly advantageous where the
first electrode compartment operates at a lower pressure than the second
electrode compartment. This differential pressure may occur where the
first electrode is an anode and the second electrode is a cathode. In one
embodiment, the substantially incompressible components include an anode
flowfield and a porous anodic electrocatalyst substrate. Particularly, the
anode compartment may comprises a flattened expanded metal flowfield and
the cathode compartment may comprise a non-flattened expanded metal
flowfield. Furthermore, the anode flowfield and porous anodic
electrocatalyst may be molded into a cell frame. The electrolyzer is
beneficial for assuring that the proton exchange membrane does not deform
under fluid pressure forcing the proton exchange membrane against the
rigid, porous electrocatalyst substrate.
Another aspect of the invention provides a hydrogen delivery system of the
vehicle comprising a controller for controlling the performance of the
system. The controller sends the command signals to cause the system to
release hydrogen and provide oxygen to the vehicle's exhaust system or
catalytic converter.
The controller sends a command to a valve or system of valves in the
delivery system to open the valve or valves. This command may be a
sequence of commands or signals that cause the valves to cycle in such a
way as to deliver hydrogen according to one of several delivery modes. The
modes may be time, volume, or converter temperature dependent. Further,
the modes may include a continuous delivery, packetized delivery, pulsed
delivery or continuously varying delivery of hydrogen.
The hydrogen delivery system may have sensors placed in appropriate parts
of the system for sensing various parameters of the system and sending
signals to the controller indicating the value of these parameters. If the
parameters fall within a certain window, then the system is working
properly. Otherwise, some error may have occurred in the system, and an
appropriate response may have to be executed, which may just present a
warning indicator, or may actually shut down the system.
The vehicle may generate one or more trigger signals to the controller to
instruct the controller to cause the release of hydrogen. Additional
signals, such as temperature sensors, may instruct the system to slow or
stop the release of hydrogen.
The controller also sends the command to the hydrogen generating part, or
electrolyzer, of the system to turn on hydrogen generation. A signal from
the hydrogen storage vessel that the vessel is full may prompt the
controller to turn the hydrogen generator off. Additional signals, such as
low battery voltage or high power demand on the engine, may temporarily
signal the system to stop electrolyzing. Examples would be to slow or stop
generation when the vehicle is at idle with the air conditioner turned on,
or when the vehicle is passing or accelerating. Likewise, the system could
be designed to increase the hydrogen production rate to take advantage of
vehicle energy when coasting or the like.
BRIEF DESCRIPTION OF THE DRAWINGS
So that the features and advantages of the present invention can be
understood in detail, a more particular description of the invention,
briefly summarized above, may be had by reference to the embodiments
thereof which are illustrated in the appended drawings. It is to be noted,
however, that the appended drawings illustrate only typical embodiments of
this invention and are therefore not to be considered limiting of its
scope, for the invention may admit to other equally effective embodiments.
FIG. 1 is a schematic diagram of a hydrogen production and delivery system
of the present invention for heating a catalytic converter.
FIG. 2 is an exploded view of a preferred electrolyzer that may be employed
in the present invention.
FIG. 3 is a schematic of an alternate hydrogen storage system of the
present invention.
FIG. 4 is a cross-sectional side view of a catalytic converter monolith
showing air and hydrogen flow in the axial direction.
FIG. 5 is a cross-sectional front view of the catalytic converter monolith
in FIG. 4;
FIGS. 6(A-D) are graphs of the catalyst temperature measured at axial
positions within the monolith as indicated in FIG. 4. FIGS. 7(A-D) are
graphs of the catalyst temperature measured at radial positions within the
monolith as indicated in FIG. 5.
FIG. 8 is a schematic diagram of a system for the combustion of hydrogen in
an internal combustion engine.
FIG. 9 is a schematic diagram of a system having oxygen recovery and
storage equipment and means for injecting oxygen into the catalytic
converter.
FIGS. 10(a-b) are graphs of the catalyst temperature at various axial and
radial distances in the catalyst monolith over a period of 50 seconds
using a pulsed release of hydrogen into an air stream.
FIG. 11 is a schematic diagram of an electrolyzer having dummy cells for
heating or cooling the electrolyzer using liquid from a vehicle radiator.
FIGS. 12(a-b) are catalyst monoliths having hydrogen injection
distributors.
FIG. 13 is a schematic diagram of a catalytic converter having isolation
valves allowing hydrogen or oxygen to diffuse evenly throughout the
monolith before delivery of another gas to provide a combustible mixture.
FIGS. 14 (a-e) are schematic diagrams of alternative electrical systems for
providing electrical power to the electrolyzer.
FIG. 15 is a flowsheet showing the modes of operation for an ideal
chemically heated catalyst (CHC) system.
FIG. 16 is a block diagram of a control system for a hydrogen production
and delivery system of the present invention.
FIG. 17 is a flow chart for the main procedure for a controller system and
on-boarddiagnostics for the present invention. FIGS. 18 is a flow chart
for the process that the control system goes through while it is waiting
for ignition of the vehicle.
FIG. 19 is a flow chart showing the process that the system goes through to
determine if hydrogen is to be released in response to a key trigger
release or a post key trigger release.
FIG. 20 is a flow chart showing the process that the system goes through
while it waits for a crank trigger signal.
FIG. 21 is a flow chart showing the process that the system goes through to
determine if hydrogen is to be released in response to a crank trigger
release or a post crank trigger release.
FIG. 22 is a flow chart showing the process by which the system produces
hydrogen through electrolysis.
FIG. 23 is a flow chart showing the procedure that the system goes through
to check whether any error flags are set.
FIG. 24 is a flow chart showing the procedure that the system goes through
in a check for a hydrogen leak from the system.
FIG. 25 is a flow chart showing the procedure that the system goes through
to release hydrogen to the vehicle's exhaust system.
FIG. 26 is a flow chart showing the procedure that the system goes through
to check certain parameters of the system prior to producing hydrogen.
FIG. 27 is a flow chart showing the procedure that the system goes through
to maintain the power to the stack of cells within the electrolyzer.
FIG. 28 is a flow chart showing the procedure that the system goes through
to monitor certain parameters during the production of hydrogen.
FIG. 29 is a flow chart showing the procedure that the system goes through
to shut down the electrolyzer and both oxygen and hydrogen delivery
equipment.
FIG. 30 is a flow chart showing the procedure that the system goes through
to release hydrogen in timed packets for a specified time period.
FIG. 31 is a flow chart showing the procedure that the system goes through
to release hydrogen in timed pulses for a specified time period.
FIG. 32 is a flow chart showing the procedure that the system goes through
to release hydrogen in timed packets for a specified volume of hydrogen.
FIG. 33 is a flow chart showing the procedure that the system goes through
to release hydrogen in timed pulses for a specified volume of hydrogen.
FIG. 34 is a flow chart showing the procedure that the system goes through
to set the current delivered to the stack of cells.
FIG. 35 is a flow chart showing the procedure that the system goes through
to periodically update the parameters in the system.
FIGS. 36 and 37 are schematic diagrams of two alternate embodiments of the
system of FIG. 1, wherein the water recovery system in the vessel is
eliminated.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
The present invention provides a method and apparatus for chemically
heating a catalyst bed by feeding hydrogen to the catalyst. The invention
also provides a method and apparatus for thermally conditioning a catalyst
in order to enhance the conversion of unacceptable emissions (emanating
from an internal combustion engine) into water and other acceptable
emissions. In one aspect of the invention, hydrogen is supplied from an
electrolyzer or other hydrogen source and injected into the monolith of a
catalytic converter to more rapidly bring the catalyst to a light-off
temperature.
In another aspect of the invention, an apparatus is provided for producing,
storing and delivering hydrogen to an automotive catalytic converter. The
apparatus is simple and compact and includes a minimal number of moving
parts. The apparatus includes a water source, an electrolyzer, a hydrogen
storage vessel, and a valve for controlled delivery of the hydrogen to the
catalytic converter. The apparatus preferably also includes an
oxygen/water separator in communication with the anode flowfield, a
hydrogen/water separator in communication with the cathode flowfield,
and/or a check valve to prevent backflow of hydrogen from the storage
vessel to the cathode flowfield. It may also be preferred to include an
oxygen source in communication with the catalytic converter to facilitate
the oxidation of hydrogen. The electrolyzer must be electronically coupled
to a power source and the electrolyzer may be operated in various manners.
In yet another aspect of the invention, the electrolyzer and water
reservoir are provided in a relationship that allows the oxygen gas
bubbles generated in the anode to rise out of the electrolyzer and into
the water reservoir and/or oxygen/water separator. The water reservoir
preferably also serves as an oxygen/water separator wherein the oxygen may
be released or stored and the water may be recycled to the anode. By
disposing the water reservoir inlet above the anode flowfield or manifold
outlet and providing a flow line that has a substantially continuous
upward slope, the natural buoyancy of the gas bubbles lifts water from the
anode and causes a natural circulation of water into and out of the anode.
This arrangement may eliminate the need for a water pump, yet provides for
water delivery and cooling to the anode and cathode.
In a further aspect of the invention, the flow line between the anode
flowfield outlet and the water reservoir inlet is equipped with a bubble
detector which indicates that there is sufficient water being supplied to
the anode. The bubble detector may be used alone or in combination with
other sensors to verify proper production, collection, delivery and
integrity of the system or its components. A similar bubble detector could
be placed between the cathode flowfield and the hydrogen storage tank.
Another aspect of the invention provides for water level management in the
system to avoid hydrogen leaks. The hydrogen storage tank preferably
includes a water level sensor in order to maintain at least a minimum
amount of water in the bottom of the vessel where various instruments and
pipes interface with the vessel. Similarly, it is preferred that the anode
and cathode manifolds be configured so that the electrolyzer is always
full of water. The presence of water assists in maintaining a seal between
adjacent components.
A further aspect of the invention provides a hydrogen delivery system
designed to supply pulses of hydrogen gas to the catalytic converter or
any other desired component of the automobile, such as the engine. The
hydrogen delivery system comprises a pair of valves with a region
therebetween having a known volume. The region is filled with hydrogen
under pressure by opening the first hydrogen valve communicating with the
hydrogen storage vessel. The first valve is then closed and the second
valve is opened to release the pressurized hydrogen into the catalytic
converter, exhaust manifold or engine, as desired. While the region may be
of any useful volume and configuration, a typical automobile will
preferably have a region suitable to contain about 5 cubic centimeters
(cc) at about 400 pounds per square inch (psi), which is released as about
100 cc at about 1 atmosphere (atm). The invention also provides for a
turnstile valve as an alternative to the hydrogen delivery system just
described. It is believed that the turnstile valve may be more reliable. A
third delivery design may consist of a solenoid value for on/off control
in series with a variable flow restriction valve for flow control.
Another aspect of the invention provides for a hydrogen flow path to be
maintained under pressure to prevent the infiltration of oxygen therein.
While minute amounts of oxygen may enter the system or small amounts of
hydrogen may leak out of the system, the invention maintains safe hydrogen
and oxygen concentrations throughout the system.
The invention also provides for the use of a low carbon-containing alcohol,
e.g., methanol, ethanol or propanol, in the water throughout the system in
order to depress the freezing point of the water. An alcohol, such as
methanol, may be provided in any useful concentration, but preferably at
an alcohol:water molar ratio of between zero and about 1:1. The alcohol is
carried with the water throughout the system and may be oxidized at the
anode to carbon dioxide gas (CO.sub.2), six hydrogen ions (H+), and six
electrons (e-). The hydrogen ions, or protons, and electrons are
recombined at the cathode to form three molecules of molecular hydrogen
(H.sub.2).
The invention further provides an electrolytic cell and method of making an
electrolytic cell having a flowfield and catalyst substrate molded
together within a common frame. While typical cell components are merely
held together under pressure between two end plates, a one piece, molded
cell would enhance the performance and extend the life of an electrolytic
cell in applications in which the cathode and anode are operated at
different pressures, particularly at high differential pressures
approaching 400 psi or more.
The invention preferably includes a check valve between the electrolyzer
and the hydrogen storage vessel. The check valve prevents the pressurized
hydrogen in the vessel from flowing back into the electrolyzer,
specifically the cathode. Therefore, if the electrolyzer has any small
leaks, the electrolyzer may depressurize down to atmospheric pressure
without a substantial loss of hydrogen.
FIG. 1 is a schematic flow diagram of a system 10 of the present invention
installed on a vehicle to deliver hydrogen to the exhaust system. The
vehicle includes a catalytic converter 12 located in an exhaust line 14
from a vehicle's exhaust manifold, as shown. The exhaust line 14 is
provided with hydrogen from a hydrogen inlet line 16 and, preferably, air
from an air pump 18. The air pump could be any suitable air source for
injecting air into the exhaust line at suitable pressure and volumetric
flow rate to achieve any desired air/hydrogen ratio mixture. The air pump
may be replaced by a venturi wherein the pressurized hydrogen gas provides
a motive force sufficient to draw in ambient air.
The system 10 includes a water reservoir 20, an electrolyzer 22, a hydrogen
storage vessel 24 and a hydrogen delivery system 25. The electrolyzer 22
may preferably comprise a plurality of stacked identical cells 26 between
end plates 28. The water reservoir 20 preferably serves not only as a
water storage chamber but also as a separator for oxygen gas and water and
means of dissipating heat. The reservoir 20 may be a vehicle's windshield
washer fluid storage container, but is preferably a dedicated
reservoir/separator, optionally allowing collection and storage of oxygen
via port 30. Water flows by gravity drain or is pumped from the reservoir
20 to the anodes of the electrolyzer 22 via a supply line 32. As the
anodes produce oxygen, the oxygen and entrained water rises naturally back
to the reservoir 20 via a return line 34.
A bubble detector 37 is preferably disposed adjacent or around the return
line 34 in order to sense the passage of gas bubbles therein. A suitable
type of bubble detector is an optical transmission type detector. The
electronic signal from the bubble detector 37 is preferably transmitted to
a controller 134 and the signal indicates that at least certain aspects of
the electrolyzer 22 are operating properly. Conversely, if the signal
indicates the absence of gas bubbles in the return line 34, then the
electrolyzer 22 may have a problem, such as a ruptured proton exchange
membrane (PEM), depleted or blocked water supply, etc. Furthermore, the
controller 134 receiving the signal from the bubble detector 37 may
analyze the number of gas bubbles passing through the line 34 over a
period of time as some indication of the electrolyzer's operating
efficiency.
The hydrogen produced at the cathodes of the electrolyzer 22 is delivered
under pressure to a hydrogen storage system 23 comprising a hydrogen
storage vessel 24 alone or in communication with one or more additional
hydrogen storage vessels 36. The use of more than one hydrogen storage
vessel is not necessary to the operation of the invention, but may be
desirable to provide sufficient storage capacity while making
accommodation for space limitations on board the vehicle. The storage
vessel may be made from various materials, such as aluminum, carbon steel,
stainless steel or a noncorrosive plastic material. It is preferred that
the hydrogen storage vessel 24 or combination of vessels 24,36 have a
total hydrogen storage capacity that is sufficient to heat up the catalyst
to a light off temperature at least one time. By storing such amount, the
rate of hydrogen production from the electrolyzer can be reduced
substantially below the rate of hydrogen consumption by the catalytic
converter during the start up period. Furthermore, the electrolyzer may be
controlled to refill the storage vessel with hydrogen at various times,
preferably when the vehicle is operating efficiently.
The system preferably also includes a water recycle line with valves 38, 39
to capture and reuse most of the entrained water from the hydrogen exiting
the electrolyzer 22, a check valve 40 to prevent back flow of hydrogen
from the storage vessel 24 into the electrolyzer 22, and a pressure relief
valve 42 to protect the system against over pressurization. Because the
storage vessel 24 may be designed to operate at a much higher pressure
than the water reservoir 20 or the supply line 32, the mere opening of the
recycle valve 38 causes water to flow out of the vessel 24. Valves 38 and
39 are alternately cycled to allow small, well controlled amounts of water
out of the cathode reservoir. As valve 38 opens, water flows into the
tubing between the valves and compresses the headspace in the vertical
stub 41. Valve 38 closes and valve 39 opens allowing the compressed
hydrogen in the stub 41 to push the captured water into the anode
reservoir. The recycle valves 38 and 39 are preferably controlled to
maintain the water level in the vessel 24. A preferred water level will be
high enough to cover most or all of the pipe and instrument couplings
along the bottom of the vessel yet not so high that the hydrogen storage
capacity of the vessel is wasted.
FIGS. 36 and 37 are schematic diagrams of two alternate embodiments of the
system of FIG. 1, wherein the water recovery system in the vessel 24 is
eliminated. Both alternate systems 220, 240 utilize a proton exchange
membrane with the cell 22 which, when operated at sufficiently low current
density, will provide the necessary reaction water to the anode when that
water is provided at the cathode. The system 220 of FIG. 36 includes a
pressurized cathode (hydrogen) reservoir 222 with means 224 for releasing
the pressure and refilling the reservoir with water. Alternatively, the
reservoir 222 could be filled using pressurized water, which would allow
manufacturing control over filling if the system requires specialized
equipment.
To eliminate the complexity and safety issues involved, the system 240
shown in FIG. 37 uses an anode reservoir 242 and a cathode reservoir 244.
This system 240 still utilizes osmotic movement of the water from the
cathode to the anode in normal operating mode but new water is provided at
the anode (at ambient pressure) and this water is transferred to the
cathode via electroosmosis during the initial operation of the
electrolyzer 22 after filling the anode reservoir 242. Once all the anode
water is depleted, electroosmosis between the reservoirs 242,244 ceases
and, once again, all process water is provided through the cathode.
Referring back to FIG. 1, the hydrogen is delivered out of the storage
vessel 24, preferably through a standpipe 44, or from the top of the
vessel 44 or 36 in accordance with the operation of a hydrogen delivery
system 25. The hydrogen delivery system 25 determines the manner in which
the hydrogen is introduced into the catalytic converter and may be as
simple as a single solenoid valve. While hydrogen may be introduced
through various valves or orifice plates as a slow, continuous stream, it
has been found (as described in relation to FIG. 10) that introduction of
hydrogen pulses or packets to the catalytic converter provides similar
heating, yet requires less hydrogen. One preferred delivery system 25 for
providing hydrogen pulses or packets comprises a first hydrogen valve 46,
a second hydrogen valve 48 and a central region 50 disposed between the
first and second valves 46,48 having a defined volume. The central region
50 is filled with pressurized hydrogen from the storage vessel 24, 36 by
opening the first hydrogen valve 46 communicating with the hydrogen
storage vessel 24, 36. The first hydrogen valve 46 is then closed and the
second hydrogen valve 48 is opened to release the pressurized hydrogen
from the central region 50 into the catalytic converter 12, exhaust
manifold 14 or engine, as desired. While the region 50 may be of any
useful volume and configuration, a typical automobile will preferably have
a region suitable to contain about 5 cubic centimeters (cc) at about 400
pounds per square inch (psi), which is released as about 100 cc at about 1
atmosphere (atm). Alternatively, the hydrogen delivery system 25 may
comprise a turnstile valve. It is believed that the turnstile valve may be
more reliable over a period of extended use.
The ignition switch 132 is preferably electronically coupled to a
controller 134 which controls the operation of the hydrogen valves 46, 48
of the hydrogen delivery system 25. In one preferred method of operation,
the hydrogen valves 46, 48 provide pulses of hydrogen to the manifold 14
until the temperature sensor 136 reads a temperature equal to or greater
than the light-off temperature. The temperature sensor provides feedback
to the controller so that adjustments in the manner of operating the
hydrogen valves 46, 48 can be made. It should be recognized that any
number of temperature sensors could be used and that the controller may
take account of any number of conditions in determining the appropriate
operation of the valves 46, 48, or other hydrogen delivery systems 25 such
as a rotary valve. It is anticipated that refinements in the hydrogen
delivery may include pulses of varying frequency and/or volume over time
to compensate for increasing catalytic converter temperature or decreasing
pressure in the hydrogen storage reservoir.
The electrolyzer 22 receives power from a source 138. It is preferred that
the electrolyzer produce hydrogen when the hydrogen pressure in or near
the hydrogen storage vessel 24, as indicated by pressure sensor 144, falls
below a setpoint pressure between about 100 psig and about 400 psig. It
should be recognized that the power to the electrolyzer 22 should be
turned off when the pressure exceeds a high pressure setpoint, such as 400
psig. It should also be recognized that many other conditions may be
considered in controlling the electrolyzer 22, such as signals from the
engine management unit, the length of time that the vehicle has been
running, the characteristics of the power supply and the volume of the
storage vessel. The other conditions mentioned may be relevant to the
efficient operation of the vehicle and the timely replenishment of the
hydrogen supply for use during the next cold start. However, in order for
the system to account for multiple conditions, it is preferred that the
system communicate with a microprocessor controller 134, whether the
controller is dedicated to the system 10 or provided as part of the
vehicle (such as the engine management controller,) or some combination of
controllers.
The microprocessor controller 134 may also be used to monitor and control
the water level in hydrogen storage vessel 24. The hydrogen storage vessel
24 is preferably equipped with a water level sensor 146, most preferably a
sensor without any moving parts such as an optical sensor. When the water
level is too high, the water outlet valve 38 is opened so that the
pressure in the vessel 24 will drive water out of the vessel into the
reservoir 20 or supply line 32. When the water level is within an
acceptable range, the valve 38 is closed. Again, it is preferred to
maintain a sufficient level of water in the vessel 24 in order to provide
additional protection against hydrogen leaks. Other various control
schemes and considerations may be employed as will be readily recognized
by those with skill in the art which are within the scope of the present
invention. For example, the microprocessor may also be programmed to carry
out timed control functions apart from responding to sensory inputs, and
may also serve various safety functions.
FIG. 2 is an exploded view of a preferred electrolyzer 22 that may be
employed in the present invention. In the following description of the
electrolyzer 22, the materials of construction referred to as "preferred"
are the materials actually used in a test device to prove that the
invention would work for its intended purpose. In commercial production
models of the present invention, where possible, less expensive materials
may be used throughout, such as carbon steel replacing titanium where
possible, and plastics, such as polypropylene, where heat and stress will
permit the use of such materials.
The electrolyzer 22 may be referred to herein as a proton exchange membrane
(PEM) electrolyzer. The proton exchange membrane 72 may itself prove
corrosive in this environment in contact with certain substances, thus
requiring the careful selection of the material of construction of the
electrolyzer. For example, the PEM 72 should only come in contact with
carbon, graphite, valve metals (such as titanium or tantalum), noble
metals (such as platinum or palladium) or gold. However, those of skill in
the art will readily recognize where less exotic materials than those
listed in the following discussion that are located away from the PEM
material itself and the oxygen electrode catalyst can be readily employed
without penalty. For example, graphite will be the material of choice in
certain structural elements, and not some obvious candidates such as
copper, aluminum, or iron, which can corrode thus forming ions that can
poison the anode and/or cathode electrocatalysts.
The electrolyzer 22 includes an anodic electrocatalyst substrate and/or
current collector 70 and a flattened "expanded" titanium flow field 68
held within an anodic cell frame 66 made of polychlorotrifluoroethylene
(PCTFE) sheet (such as KEL-F available from the 3M Company, St. Paul,
Minn.). The preferred anode substrate and/or current collector is a thin
sheet of porous titanium made by sintering small diameter titanium spheres
and is available from AstroMet, Cincinnati, Ohio. A more preferred anode
substrate and/or current collector is a thin sheet of porous titanium made
by sintering small diameter titanium fibers and is available from Porous
Metal Products, Jacksboro, Tex.
The electrolyzer 22 further includes a cathode substrate and/or current
collector 76 and an expanded stainless steel flow field 78 retained in a
cathodic cell frame 80 formed of polysulfone. The preferred cathode
substrate and/or current collector is a carbon paper consisting of pressed
carbon fibers or a carbon cloth made from a weave having a cathodic
electrocatalyst layer on one side containing polytetrafluoroethylene
(PTFE)-bonded high surface area colloidal platinum or palladium, supported
on carbon black or preferably an electrolessly deposited or electroplated
thin film of platinum or palladium, most preferably having a platinum or
palladium loading of at least 0.1 mg/cm.sup.2. Alternatively, the cathode
may be constructed using a semi-compressible stainless steel felt,
suitably supported, having high porosity and sufficient catalytic
activity.
The various components of the PEM electrolyzer are stacked together and
retained with a plurality of tie rods 82, preferably 16 such tie rods.
Stainless steel tubes, such as SS316, are then screwed into four threaded
ports on one of the titanium end plates. The ports are the anode water
inlet port 56, the anode water/oxygen outlet port 58, and a pair of
cathode hydrogen/water outlet ports 84. To minimize electrical contact
resistance between components, the titanium end plates 60 and 62 and the
expanded titanium metal current collectors 68 and 78 may be electroplated
with a thin film of gold or noble metals, such as platinum.
The cathode and the anode of the electrolyzer are of special construction.
The cathodic electrode structure for hydrogen evolution may be fashioned
from a commercially available fuel cell gas diffusion layer on a carbon
cloth backing (such as ELAT available from E-TEK, Inc., Natick, Mass.),
which acts as a support for the active hydrophilic electrocatalyst layer.
This active layer contains high surface area colloidal platinum (about 100
m.sup.2 /g), supported on carbon black (between about 10 and about 50 wt %
Pt on C), yielding a platinum loading of at least about 0.1 mg/cm.sup.2.
The cathodic electrode structure may be hot-pressed onto one side of a
segment of a precleaned PEM material. Hot-pressing of the cathodic
electrode and PEM is preferably carried out between the plates of a
hot-press elevated to about 200.degree. C. for about 60 seconds, and using
a force of about 15,000 pounds.
One suitable anodic electrocatalyst layer contains mixed iridium and
ruthenium dioxides at a molar ratio of about 1:1. The layer is prepared by
dissolving iridium and ruthenium chlorides in about 8 ml of concentrated
HCI and heating the mixture to almost dryness. The resulting chlorides are
then dissolved in isopropanol to make an ink-line coating. A porous
titanium plate (such as a 0.05" thick plate available from Astro Met of
Cincinnati, Ohio) is etched in 12% HBF.sub.4 for about 60 seconds and
rinsed with isopropanol. This substrate is then coated with the ink-like
mixture and the solvent evaporated under low heat of about 90.degree. C.
This coating and drying procedure is repeated several times, then the
electrode is heated in a furnace at 400.degree. C. for 10 minutes in
ambient air. The coating, drying, and furnace treatment is repeated twice
more, but with a final baking time of two hours instead of 10 minutes. A
preferred anodic electrocatalyst consists of high surface area
platinum-ruthenium metal alloy powder having an atomic ration of 1:1,
platinum:ruthenium and which is available from E-TEK, Inc., Natick, Mass.
A more preferred anodic electrocatalyst consists of high surface area
platinum-ruthenium oxide (Pt-RuO.sub.x) powder having an atomic ratio of
1:1, platinum:ruthenium. The synthesis of the Pt-RuO.sub.x electrocatalyst
employs anhydrous chloride salts of platinum and ruthenium. Quantities of
the two salts are utilized such that the Pt-Ru atomic ratio is 1:1. The
salts are ground together with NaNO.sub.3 using a mortar and pestle until
a homogenous powder is obtained. The resulting mixture is then deposited
in a crucible, placed into a 500.degree. C. oven, and fired in ambient
atmosphere for 10 minutes. After heating, the crucible is removed from the
oven and allowed to cool to room temperature. A shiny black crystalline
solid is observed in the crucible which consists of the desired
Pt-RuO.sub.x product dispersed in a salt matrix. To dissolve the salt and
liberate the insoluble catalyst product from the matrix, the crucible is
placed into a beaker of distilled/deionized water for 24 hours. The
product is then isolated using vacuum filtration, washed with copious
amounts of deionized water, and dried in a heated vacuum oven.
To ensure that surface atoms of platinum and ruthenium in the Pt-RuOx
crystallites are activated completely, the catalyst is subjected to a
heated reduction step. The catalyst is placed into a cool ceramic heating
tube which is housed in a muffle furnace and equipped for the external
supply of gaseous reactant. Initially, argon is allowed to flow for one
hour in the tube to remove potential oxidizing species from the ceramic.
Hydrogen is then introduced into the ceramic tube and the temperature of
the furnace/ceramic tube is increased slowly to 200.degree. C. This
environment is maintained for six hours. The resulting catalyst is a very
fine powder which displays catalytic activity toward methanol oxidation in
the presence of air.
An even more preferred anodic electrocatalyst consists of a homogenous
mixture of high surface area iridium and ruthenium oxides with either high
surface area platinum-ruthenium metal alloy powder or high surface area
Pt-RuO.sub.x. In each case, the mole ratio of iridium and ruthenium oxides
to platinum-ruthenium metal alloy or to Pt-RuO.sub.x. should preferably be
1:1.
The anodic eletrocatalysts described above are applied to one side of a
proton exchange membrane in the form of a catalyst ink that is prepared
using the following procedure. The catalyst is first dispersed into
distilled water using a water:catalyst ratio of approximately 1:1 by
weight. The mixture is then sealed into an appropriate container to
prevent solvent evaporation and agitated using high energy sonication for
one hour or more until complete dispersion is achieved. Following
dispersion and wetting of the catalyst, a quantity of
commercially-available, dissolved ionomer, solution, such as the 5 wt %
Nafion.RTM. solution available from Solution Technologies, Mendenhall,
Pa., is added to the dispersion such that the final concentration of
ionomer is approximately 15 wt %.
This mixture is resealed into a container to minimize evaporation and
subjected to high energy sonication until a homogenous, well-dispersed
catalyst ink is achieved. The ink is applied directly to the surface of a
dry membrane as a single layer or as a multilayer coating using a brush
technique. For multilayer coatings, the solvent of the previously applied
layer is evaporated to dryness prior to the application of the next coat.
The procedure could be modified to make use of other coating application
techniques such as air spraying or spin coating which are more amenable to
mass production.
After the catalyst ink has been applied successfully to one side of the
membrane the membrane and electrode (M&E) assembly is subjected to a hot
press step. In this step, the electrodes and membranes are fused under
elevated temperature and pressure in a hydraulic press equipped with
resistively-heated platens. To press the M&E, a "press package" is created
which consists of the M&E and various insulative and support layers to
facilitate the release of the M&E from the package after the press step.
In this package, the M&E assembly is sandwiched typically between two thin
Teflon sheets to protect the electrode surface(s), and this sandwich is
placed between two thin, flat metal plates that provide support and heat
transfer between the platens and the M&E during the press step. The
complete M&E, Teflon, and metal plate press package is positioned quickly
between the two pre-heated platens (ca. 150 to 200.degree. C.), and
pressure (1000 to 1500 psi) is applied to the package for 60 to 120
seconds to finish the M&E fabrication process.
For the longest component lifetime it is important to prevent the movement
of internal electrolyzer components as the cathode (hydrogen) pressure is
changed. This is accomplished by selecting an anode flowfield and a porous
electrocatalyst substrate that are noncompressible and selecting cathode
components which provide sufficient elasticity such that the PEM is
pressed firmly against the anode regardless of the cathode pressure.
In the preferred structure the anode flowfield consists of flattened
"expanded" metal and the cathode flowfield includes at least one piece of
non-flattened (as expanded) metal to provide compression.
An alternative flow field would be perforated corrugated steel. The
corrugates could be designed to provide more rigidity on the low pressure
side and more elasticity and resiliency on the high pressure side of the
membrane.
In mass production, where components such as the cell frames are molded,
the anode flowfield components could be molded into the cell frame. This
would increase the rigidness of the noncompressible side as well as
simplify assembly.
FIG. 3 is an alternative hydrogen storage system 25 which utilizes a metal
hydride type storage vessel 52. Prior to operation, the system 25 of FIG.
3 permits purging all air from the storage system with an inert gas, such
as nitrogen, by attaching a nitrogen gas feed line at a purge gas inlet 54
downstream of the check valve 40. During the purging operation, the metal
hydride vessel 52 is detached at a quick disconnect 86. This operation
effectively seals both the vessel 52 and a gas line 88, to keep the purge
gas out of the vessel 52. The remainder of the system 25 is then purged
from the purge gas inlet 54 through a back pressure regulator 90.
To charge the system 25 with hydrogen, a needle valve 92 between the
storage vessel 52 and the back pressure regulator 90 is shut. Hydrogen gas
generated by the electrolyzer (See FIG. 1) is preferably processed through
a four-stage process to remove entrained water (liquid or vapor) and any
oxygen contaminant from the hydrogen stream before storage in the vessel
52. The first step involves removal of a small amount of entrained liquid
water coming from the electrolyzer in the hydrogen gas. The entrained
liquid water is removed without a pressure loss by means of the entrained
liquid water trap 94. The second step involves cooling the hydrogen gas
stream from the electrolyzer temperature to ambient in a condensing coil
96. The electrolyzer typically operates at between about 20 and about
60.degree. C. above ambient, with the exact temperature depending on
specific electrolyzer operating conditions. This second step condenses a
substantial portion of the water vapor in the hydrogen gas stream. This
condensed water could absorb a significant amount of alcohol, which may be
present during operation using windshield washer fluid as the electrolyzer
reactant feed. The condensate is collected in a condensate collector 98
and removed through the drain valve 38.
At this point, the hydrogen gas stream is still saturated with water vapor,
but now at a lower temperature. This saturated gas stream is next passed
into a zeolite-filled gas drier 100. This drier absorbs water vapor and
any alcohol vapor present when using a windshield washer fluid feed. Any
oxygen contaminant present in the hydrogen gas stream is then eliminated
in a catalytic recombiner or oxygen eliminator 102 to reduce it to water.
Final clean-up of the hydrogen gas stream is accomplished in a second
zeolite absorber bed in a polishing drier 104. The polishing drier removes
traces of water produced by the catalytic recombiner 102.
The hydrogen storage system 25 of FIG. 3 is designed for relatively short
term operation. The system 25 may be designed for longer term operations,
for example 100,000 miles or more, utilizing other methods of water
removal known in the art. A satisfactory metal hydride hydrogen storage
unit is available from Hydrogen Consultants of Littleton, Colo. Such an
available unit can store 30 liters of hydrogen which can be delivered at
30-45 psig, with recharging using hydrogen gas at 100-200 psig. Still, the
most preferred hydrogen storage means is a pressure vessel, such as vessel
24 of FIG. 1, made of a composite structure, involving the use of aluminum
or ferrous-based alloys. A suitable hydrogen storage vessel of this type
is available from Harless Specialties of Irwin, Pennsylvania.
FIG. 4 is a simplified cross-sectional view of a catalytic converter
monolith showing air and hydrogen flow in the axial direction through the
monolith 13. The temperature of the monolith is measured with a
thermocouple at points 31(a)-(e) along the central axis, with point 31(a)
being on the front face where the gases first contact the catalyst and the
other points 31(b)-(e) located at positions successively further into the
monolith. The results of these temperature measurements at 40 liters per
minute (1 pm) total gas flow rate containing 3%, 5%, 8.5% and 17% hydrogen
is shown in FIGS. 6(a)-(d). FIGS. 6(a)-(d) are graphs of the catalyst
temperature measured at axial positions within the monolith as indicated
in FIG. 4.
Now referring to FIG. 5, a simplified cross-sectional view of the catalytic
converter monolith 13 of FIG. 4 is presented. The temperature of the
monolith 13 is measured with a thermocouple at points 33(a)-(c) along the
monolith radius, with point 33(c) being in the center of the monolith and
the other points 31(b) and (a) located at greater distances from the
center. The results of these temperature measurements at 40 liters per
minute (lpm) total gas flow rate containing 3%, 5%, 8.5% and 17% hydrogen
is shown in FIGS. 7(a)-(d). FIGS. 7(a)-(d) are graphs of the catalyst
temperature measured at radial positions within the monolith as indicated
in FIG. 5.
The face of the monolith should be noted as the beginning of the active
catalyst sites, not the physical front of the brick. Using this
definition, the face will move as the catalyst ages.
It has been found that the introduction of a relatively small percentage of
hydrogen in an air stream within a typical automobile exhaust system
provides nearly spontaneous heating of a major portion of a face 35 (see
FIG. 4) of the catalyst material almost immediately following ignition in
the internal combustion engine providing the exhaust gas. This heating
along the face 35 of the converter is fortuitous because it has been found
that the most effective site for providing local heating is along and near
the upstream face 35 of the catalyst monolith 13. In fact, where the
monolith 13 is made of a material that heats slowly when used in
association with the present invention, the face 35 may comprise a more
reactive catalytic material to bring the entire catalytic converter to
light-off more quickly. In addition, the heat supplied by the spontaneous
combination of hydrogen with oxygen or air in the presence of the catalyst
monolith 13 produces only a small quantity of water as a product of the
reaction, which does not degrade the performance of the catalytic
converter.
The air flow rate through the monolith 13, depending on engine size and
tuning parameters, typically falls in the range between about 40 and about
400 liters per minute (lpm). The ideal range is between 80 and 300 lpm,
depending on engine size. Effective concentrations of hydrogen for these
flow rates are between about 1 and about 28 volume percent, with a
preferred range between about 5 and about 18 volume percent. The most
preferred range of hydrogen concentration, again depending on engine size,
is between about 8 and about 15 volume percent. For example, at an air
flow rate of about 150 lpm across the catalytic converter, the ideal range
of hydrogen concentration in that flow is between about 12 and about 13
volume percent. Under those conditions, light-off temperature at the face
35 is reached in about one second. At an air flow rate of about 90 lpm and
a hydrogen concentration between about 8.5 and about 11 volume percent,
light-off is achieved in about two seconds.
The energy consumption to heat the catalyst varies depending on the air
flow rate and the concentration of hydrogen. For example, at an air flow
rate between about 30 and about 50 lpm and a hydrogen concentration
between about 10 and about 111/2 volume percent, the chemical energy
required to heat the monolith to light-off is approximately 1.5
Watt-hours. An electrically heated catalyst (EHC) unit requires between
about 10 to about 15 Watt-hours to heat the same monolith at the air flow
rate of 30 to 50 lpm.
The present invention is also suitable for use in low ambient temperature
conditions, for example where the outside temperature is as low as
-7.degree. C. or lower. Depending on the active catalyst compositions
used, these extremely low temperatures may cause the amount of time
required to achieve light-off to double. In those conditions, it may be
desirable to add a small electrical heater, which would be much smaller
than an EHC heater and require only about 200 Watts of power, in order to
achieve the results similar to those achieved at typical ambient
temperatures.
Now referring to FIG. 8, one aspect of the invention provides an on-board
hydrogen ignition assist system 110. A source of hydrogen, such as the
electrolyzer described above or any suitable means, fills the hydrogen
storage cylinder 24. An ignition supply line to a control valve 122
controls the supply of hydrogen into an engine ignition 124. The engine
ignition 124 includes the fuel, air, and electrical components for an
internal combustion engine 126. Thus, the hydrogen can be supplied at any
convenient location so that it is injected into the cylinders of the
engine 126. For example, hydrogen under pressure can be supplied to the
intake manifold where there is already a fuel/air mixture (during the
inlet cycle), or the hydrogen can be mixed with air before it goes to the
engine's fuel injection system, or other means.
The system 110 of FIG. 8 turns the internal combustion engine 126 into a
hydrogen fuel injected engine for the first few seconds of start-up,
before any gasoline is introduced into the engine. This way, the catalytic
converter can be brought to light-off temperature before the engine begins
producing undesirable emissions. Then, when gasoline is finally injected
into the system, the catalytic converter has already been heated to a more
efficient operating temperature, preferably the light off temperature.
Expended fuel gases are collected in an output manifold 128 and flow into
the exhaust line 14. An ignition controller 130 provides control signals
to the control valve 122 for the introduction of hydrogen to the engine
ignition 124 to coordinate hydrogen introduction during cold start
operations. The on-board hydrogen ignition assist system 110 functions
with or without the hydrogen delivery system 25 for introducing hydrogen
directly into the catalyst monolith, but the system 110 is preferably used
in combination with hydrogen delivery to the catalyst. When using both
systems 25, 110, the hydrogen generation and on-board storage can be used
for both.
Another aspect of the invention provides oxygen recovery from the
electrolyzer. FIG. 9 is a schematic diagram of a system 140 providing
equipment for oxygen recovery, storage and injection into the catalytic
converter. The oxygen separated from water in the water reservoir 20,
passes through the port 30 and is collected in a storage vessel or
cylinder 55. During ignition, and perhaps during all operation of the
vehicle, the oxygen may be released from the vessel 55 by opening a valve
57 and input into the intake of the air pump 18. In this manner, the
oxygen enriches the air and provides more efficient combustion of the
hydrogen or exhaust gases within the catalytic monolith 13. It should be
recognized that the pump 18 of FIGS. 1 and 9 may be replaced with a
venturi for drawing air into the catalytic converter.
Now referring to FIGS. 10(a) and (b), graphs are provided showing the
catalyst temperature at various axial distances and radial distances in
the catalyst monolith over a period of 50 seconds using a pulsed release
of hydrogen into an air stream. The graphs show the temperature rise in
the catalytic converter monolith at an air flow rate of 90 lpm and pulsed
hydrogen flow controlled by a microprocessor. The pulsed hydrogen flow was
provided by opening the hydrogen release valves 46 and 48 (see FIG. 1) for
0.01 seconds and closing the valves for 0.66 seconds, successively 10
times.
Comparing the temperature profiles of FIG. 10(a) with those of FIGS.
6(a)-(d) and the temperature profile of FIG. 10(b) with those of FIGS.
7(a)-(d) it is shown that light-off temperatures of between about
400.degree. C. and about 600.degree. C. can be readily attained even with
pulsed hydrogen flow. One advantage of pulsed flow is the better
utilization of the hydrogen and therefore conservation of the hydrogen
supply.
FIG. 11 is a schematic diagram of an electrolyzer 150 having non-gas
producing cells 152 used for heating or cooling the electrolyzer using
liquid from a vehicle radiator 154. These cells 152 may be comprised of
plates providing a passage for the radiator fluid between electrolytic
cells 156 in order to absorb or deliver heat to the electrolyzer. Because
electrolyzers operate most efficiently at elevated temperatures, the
radiator may be used to warm the electrolyzer in cold weather conditions.
Alternately, the radiator fluid may be used to cool the electrolyzer after
an extended period of use.
FIGS. 12(a) and (b) are two catalytic converters 12 having different
hydrogen injection means 160. In each of the figures, the converters 12
have multiple monoliths 164 separated by a region 166 for hydrogen
introduction and diffusion. In FIG. 12(a), the injection means 160 is
primarily external to the converter 12 with a plurality of injection tubes
162 delivering hydrogen into the gaps 166. Conversely, in FIG. 12(b) the
injection means 160 passes through the monoliths 164 with a plurality of
holes or ports 168 for hydrogen delivery into the gaps 166. In a very
similar manner, air or air/oxygen may be introduced in multiple locations
to improve mixing, distribution, or hydrogen and oxygen combination
characteristics such as uniformity across the face of the catalyst.
FIG. 13 is a schematic diagram of a catalytic converter 12 having an
upstream isolation valve 170 and a downstream isolation valve 172. When
the valves 170, 172 are closed, hydrogen can be injected into the monolith
13 and allowed to diffuse evenly throughout the monolith. After only a
fraction of a second for diffusion, the valve 172 is opened and oxygen is
delivered to provide a combination mixture. Alternatively, because oxygen
is a larger molecule and diffuses more slowly, it may be similarly
advantageous to first diffuse oxygen into the isolated monolith, then
introduce hydrogen while opening valve 172.
Now referring to FIG. 14(a)-(e) which are schematic diagrams outlining four
possible topologies for the powering of the electrolyzer system 200. In
FIG. 14(a) the primary source of electrolyzer power is drawn directly from
the vehicle battery 201 as well as from the alternator 203. FIG. 14(b)
eliminates the electrical draw on the battery 201 by placing diode 202
between the alternator 203 and the battery 201. Diode 202 allows current
flow from the alternator to the battery and other vehicle loads 205 but
stops current flow from the battery to the electrolyzer 200. the current
limiting circuit 206 protects the electrical system from over currents
that could be drawn by the electrolyzer as the electrolyzer resistance
changes. FIG. 14(c) is shown having an alternator 204 having an additional
winding in which the magnetic circuit provides current limiting to the
electrolyzer 200. This second winding would also allow higher voltages to
be delivered to the electrolyzer 200, allowing the number of cells within
the stack to be increased. FIG. 14(d) shows a system in which the vehicle
alternating current is drawn from the alternator 207 before the vehicle
regulator 209 and a separate current control/regulator 208 provides
electrical power to the electrolyzer 200. This topology is able to current
limit the electrolyzer load and provide higher voltages to the
electrolyzer while using a conventional alternator.
FIG. 14(e) shows a method of controlling the electrolyzer current while
maintaining overall electrical conversion efficiency. In this mode of
operation, the full potential of the vehicle's electrical system is placed
across the electrolyzer 200. Individual cells are then bypassed by
external mechanical or solid state switches 210, to lower the effective
resistance of the electrolyzer, in steps, until a current close to the
desired current is achieved. Cells may be switched in and out as desired
to maintain the current within a window.
The apparatus of the present invention, described above, may be operated in
any number of ways as will be recognized by those in the art. One
preferred mode of operating a system of the present invention, including
an electrolyzer, is described in FIG. 15. FIG. 15 is a flowsheet showing
the modes of operation for an ideal chemically heated catalyst (CHC)
system.
Control System and On-Board Diagnostics
The Control System and the On-Board Diagnostics are illustrated in FIG. 16.
The controller 134 may be the vehicle's existing built-in controller
system, i.e., engine management unit, or it may be a separate
microprocessor controller system, or even just a state machine. At a
minimum the controller 134 would need to have a processing unit (state
machine), I/O pins or ports, and a memory. The memory system may be an
EEPROM or other suitable system. The controller 134 may also have various
registers for storing data.
A preferred controller 134 has multiple outputs for sending control signals
to operate various parts of the hydrogen delivery system 10 (shown in FIG.
1), such as the hydrogen release subsystem 25, the hydrogen storage
subsystem 23, and the hydrogen generation subsystem 20, 22, 138. The
controller 134 also has multiple inputs for receiving sensor information
and other signals to control its own operation. These inputs may actually
go into controller 134, or they may be distributed to the engine
management unit, any other controllers or any combination of controllers
in the vehicle as appropriate. The control outputs from controller 134
include lines 300, 302 to dual valve subsystem 46, 48 for controlling the
turning on and turning off of the valves 46, 48, thus controlling the
delivery of hydrogen to the exhaust manifold 14, catalytic converter 12 or
engine as desired. Controller 134 also has a control output line 304 to
the electrolyzer power source 138 for controlling the supply of power to
the electrolyzer 22.
The controller 134 has control output lines 306, 307 to the valves 38, 39
in order to control when to open and close valves 38, 39 in order to
recycle the water in storage vessel 24. Controller 134 also has a sensor
input line 308 from water level sensor 146. The input from water level
sensor 146 tells controller 134 to cycle valves 38, 39 when the water
level is too high and the water needs to be removed from the cathode and
recycled to the anode. Alternative means of water recovery would include a
float valve, electroosmosis, wicking, or operation of the stack with the
anode dry and water provided by the cathode.
Controller 134 has a sensor input line 310 from hydrogen pressure sensor
144. A high or low signal from sensor 144 will tell controller 134 when to
turn off or on the electrolyzer 22 through control output line 304 to
electrolyzer power source 138. The controller 134 also has a sensor input
line 312 from bubble detector 37, an input signal line 314 from ignition
switch 132, and an output control line 316 to air pump 18 to control the
turning on and off of air pump 18.
The electrolyzer 22 may be provided with multiple performance verifications
and abnormal performance indicators for on-board diagnostics. Some of
these sensors are not shown in FIG. 1 for ease of viewing the drawing and
because a skilled person will readily understand where these sensors need
to be placed. However, FIG. 16 shows these elements in an abstract block
diagram. Controller 134 may be used to collect, store and interpret the
data from these sensors and performance indicators and take appropriate
corrective action including shutting the system 10 down and/or warning the
driver. When these operating parameters fall within acceptable ranges they
indicate that the production and storage of hydrogen is proceeding
normally. The importance of some of these parameters, and the devices and
methods used to determine some of the parameters, and the acceptable
ranges for some of them will now be discussed.
Conductivity sensors 317, perhaps two independent sensors placed at
different heights within the anode reservoir, are used to measure the
conductivity of the anode water and send a signal indicative thereof on
sensor input line 319 to controller 134. These sensors will also establish
the water level since a dry conductivity sensor shows abnormally high
resistance.
A cell voltage sensor 321 may be placed in electrolyzer 22 to measure the
voltage drop across each cell or stack of cells 26 and send a signal
indicative thereof on sensor input line 323 to controller 134. A stack
voltage measurement may provide the controller 134 with the voltage drop
occurring across all of the cells within the electrolyzer stack 26. Under
normal operation, this voltage will fall within a narrow range that is
characteristic of the electrolyzer 22, the temperature of the stack, the
stack current, and the anode feed fluid (e.g., alcohol concentration).
Stack current sensor 325, like cell voltage sensor 321, may be placed in
electronic communication with electrolyzer 22 to measure the current
through the stack of cells 26 and send a signal indicative thereof on
sensor input line 327 to controller 134. The stack current may be measured
for both performance verification as well as feedback for limiting the
maximum current demand of the electrolyzer 22.
Anode gas generation may be indicated by a bubble detector 37 in the anode
return line 34 to the water storage reservoir 20. This bubble detector
should indicate gas generation within a window related to the current.
Additionally, this detector will detect any gas leakage from the cathode
to the anode when the electrolyzer is off.
A cathode pressure sensor 329 may be placed in fluid communication with the
cathode of electrolyzer 22 to measure the pressure of the gas at the
cathode and send a signal indicative thereof on sensor input line 331 to
controller 134. The change in cathode pressure may provide a direct
reading of the hydrogen production rate.
A cathode water recovery system 38, 39, 146 may quantify the amount of
water passing through the membrane in electrolyzer 22 from the anode to
the cathode due to electroosmosis. The operation of the two valves 38, 39
will recover a known volume of water that controller 134 could keep track
of to determine whether the volume of water that has passed through
electrolyzer 22 is within acceptable limits.
An electrolyzer temperature sensor 333 may be placed within electrolyzer 22
to measure the temperature of electrolyzer 22 and send a signal indicative
thereof on sensor input line 335 to controller 134. An electrolyzer, or
stack, temperature measurement may be an abnormal operation indicator that
will typically fall within a range around the ambient temperature during
normal operation. Abnormal operation of the electrolyzer will result in
extreme temperature readings, such as temperatures greater than about
100.degree. C.
An ambient hydrogen detector 337 may be placed near the delivery system 10
to detect the presence of abnormally high levels of hydrogen in the
ambient atmosphere and send a signal indicative thereof on sensor input
line 339 to controller 134.
Improper readings in any of these measurements may indicate more than one
possible failure mode having differing effects on safety and performance
as will now be described.
A loss of water in the anode will stop hydrogen production, but should not
cause any permanent damage to the electrolyzer 22. Such a loss of anode
water would normally be indicated by the cell or stack voltage measurement
being elevated or current reduced. The loss of anode water could also be
indicated by an increase in the stack temperature measurement, or by an
indication from the bubble detector 37 of a lack of anode gas bubble
generation.
A contamination of the anode water may decrease the efficiency of the
electrolyzer and result in higher cell voltages and reduced hydrogen
production. Conductivity sensors 317 may indicate contaminated water by a
change in or improper level of conductivity.
A small leak in the proton exchange membrane may risk a combination of
hydrogen and oxygen in the anode reservoir as well as an increase in the
amount of time necessary to produce a given quantity of hydrogen. The
ambient hydrogen detector may directly detect the leak of hydrogen if that
leak is to outside the electrolyzer or sense the hydrogen coming from the
anode reservoir if the hydrogen leak is internal to the electrolyzer. An
increase in bubble production in the anode is also indicative of a
hydrogen leak as hydrogen escapes through the anode.
A rupture of the membrane may cause a venting of hydrogen gas and should
require a shut-down of the hydrogen delivery system 10 to prevent further
hydrogen release, which could lead to undesirable concentrations of
hydrogen and oxygen. The rupture and leak may be determined by the ambient
hydrogen detector 337, a hydrogen pressure drop or low rate of pressure
increase in the storage vessel 24, and possibly a reduced cell voltage.
An electrolyzer mechanical leak may leak hydrogen to the ambient atmosphere
and increase the time required to generate hydrogen. This leak may be
determined by the ambient hydrogen detector 337 or a low rate of pressure
rise in the storage vessel 24. Any leak situation will result in the
venting of hydrogen into the air where it will readily disperse. The
hydrogen storage tank is at elevated pressure, thereby preventing oxygen
from leaking into the tank.
A restriction in the flow of water through the anode may limit the
production of hydrogen due to overheating. This flow restriction may be
indicated by a decrease in gas bubble generation detected at the bubble
detector 37 or by a rise in temperature measurement in the electrolyzer
22.
Freezing of the electrolyzer 22 may slow the rate of hydrogen production
until the electrolyzer 22 can be thawed. A sufficiently low electrolyzer
temperature measurement may indicate freezing, but so may an elevated cell
voltage measurement.
A reversal of polarity between the anode and cathode may potentially result
in the release of hydrogen, from the system 10. Shut-down of the hydrogen
delivery system 10 may be required to prevent the release. The ambient
hydrogen detector 337 may indicate this failure, as will the cell voltage
measurements, the current measurement and the bubble detector since twice
as much hydrogen gas will be produced.
The quantity of stored hydrogen may be measured with any suitable pressure
sensor based on the fixed volume of the vessel 24. One good sensor is a
silicon strain type pressure sensor. The hydrogen pressure sensor 144
provides the primary data for triggering the shutdown of the electrolyzer
22 when the hydrogen storage vessel 24 is full. The controller 134 is
preprogrammed with the desired maximum pressure which it compares with the
signal received from sensor input line 310. The sensor 144 also provides
the data for triggering the closing of the valves 46, 48 when releasing
hydrogen on a volume basis during cold start, because the controller 134
may calculate the volume released from the change in pressure. The volume
release methods will be described later.
For data from the hydrogen pressure sensor 144 to be accepted as reliable,
the hydrogen delivery system 10 must provide predictable responses to
several conditions. For example when no hydrogen is being released from
the vessel 24, hydrogen generation should result in a steadily increasing
pressure reading in the hydrogen pressure sensor 144 at a rate
proportional to the measured stack current. Controller 134 may collect
this data and make this comparison.
Additionally, cathode water recovery is performed by transferring fixed
quantities of water from the hydrogen storage vessel 24 to the anode water
reservoir. Therefore, during each transfer cycle, as the water is removed,
the available volume for the hydrogen gas increases, resulting in step
drops in pressure, which can be counted by controller 134 to determine the
volume of water removed.
Hydrogen released from the hydrogen reservoir or storage vessel 24 during
delivery of hydrogen to the vehicle's exhaust system provides the simplest
test of whether the delivery system 10 is functioning properly. In other
words, hydrogen pressure must drop when the valves 46, 48 are opened, or
cycled.
Ambient temperature swings during times when the delivery system 10 is idle
provide an opportunity to measure the linearity of the hydrogen pressure
sensor 144 since the hydrogen gas follows the ideal gas law. Controller
134 may keep track of corresponding temperatures and pressures and
calculate the expected changes.
A drift in absolute accuracy of the hydrogen pressure sensor 144 could
possibly result in an over pressure or under pressure of the hydrogen. It
is the over pressure that could have catastrophic effects. The drift in
accuracy could also result in a proportional drift in the charge and
discharge of the reservoir.
A drift in the linearity in the readings of the pressure sensor 144 has the
possible effect of an over release of hydrogen as well as an over
pressure. It could also result in an under release of hydrogen, but that
is just a performance problem, rather than a safety problem. Additionally,
there could be a corresponding drift in the charge and discharge of the
reservoir 24. This linearity drift may be detected by measuring an
unexpected time period required to charge or discharge the storage vessel
24. There would also be an incorrect temperature response of the pressure
sensor 144 along the lines of the above described ambient temperature
swings.
A total electronic failure of the hydrogen pressure sensor 144 could have
the possibility of a hydrogen over pressure, since it is the pressure
sensor that provides the primary data for shutting down the electrolyzer
22. Such a failure could result in the complete loss of operation of the
hydrogen delivery system 10 even though the signal from the sensor is
within normal range. When the sensor 144 indicates no change in the
hydrogen pressure under conditions that should produce a change in
hydrogen pressure (such as a release or generation of hydrogen) then total
electronic failure of the sensor 144 may have occurred.
A mechanical failure of the hydrogen pressure sensor 144 could lead to a
hydrogen leak to the ambient atmosphere along with a complete loss of
operation of the system 10. The pressure signal may remain within a normal
range when it should be abnormal, and this would be detected as a fairly
stable hydrogen pressure reading during hydrogen generation or delivery
when there should be a significant change.
During CHC operation, or hydrogen delivery, hydrogen may be released into
the vehicle's exhaust system and the catalytic converter via two valves
46, 48, e.g., solenoid valves, operated by the CHC control system, or
controller 134. In this embodiment, the valves are in series, so both must
be opened, either together or alternatingly, before hydrogen is released.
Performance verification is concerned with the complete opening, closing,
and response time of both valves 46, 48. Hydrogen pressure stability in
storage vessel 24, as shown by the hydrogen pressure sensor 144, during
idle times may provide the leak rate of both valves 46, 48 in series or of
each valve individually if the other valve is left open. An expected
hydrogen pressure drop in storage vessel 24 during continuous opening may
indicate that each valve 46, 48 is opening as designed. The hydrogen
pressure drop may also quantify the response time of each valve 46, 48.
The controller 134 may be set for the valves 46, 48 to typically respond
to a 25 ms open pulse during which they release a fairly predictable
amount of hydrogen gas. Therefore, a significant unexpected drop in
hydrogen pressure during gas release would be indicative of a delay in
valve response time in either opening or closing.
If either valve 46, 48 is stuck closed, then there is a total failure of
the hydrogen delivery system 10, but this failure alone does not present a
safety problem, because with no drop in hydrogen pressure, there is no
trigger for turning on the electrolyzer, and the system 10 stays in a
static state.
If one valve 46, 48 is stuck open, then there is a potential for over
release of hydrogen, but the hydrogen delivery system 10 may remain
functional in some delivery modes. An unexpected hydrogen pressure drop in
vessel 24 may be indicative of this failure.
If both valves 46, 48 are stuck open, then all of the hydrogen will be
released into the vehicle's exhaust system. The system 10 will work one
final time, albeit inefficiently, and then fail completely. Total pressure
loss in storage vessel 24 is a good indicator of this type of failure as
is the temperature of the converter and response from the oxygen sensor(s)
in the exhaust system.
If a valve 46, 48 is merely slow to respond, then there is the potential
for over release of hydrogen. There is also a reduced accuracy during
pulsed release, described below. Unexpected pressure drop in storage
vessel 24 during pulsed release is the most obvious indicator of this type
of failure as is overheating of the catalytic converter.
The existing oxygen and lambda sensors may also provide feedback to the
control system. The exhaust face of the sensor contains a highly catalytic
surface, such as porous platinum, which will react with the hydrogen and
oxygen during the preheating period. The response from the sensor will be
a combination of effects including direct heating of the sensor bringing
it to operating temperature, reducing the oxygen partial pressure in the
exhaust stream due to combination with the hydrogen, and third, reduction
of the sensor output voltage due to the presence of the hydrogen in the
exhaust stream.
The use of the existing onboard oxygen sensor would allow closed loop
control of the CHC system without requiring any additional sensors in this
corrosive environment. The oxygen sensor, which is generally heated to
operating temperature by the exhaust gas, produces a signal corresponding
to the oxygen content within the exhaust gas. A cold oxygen sensor,
however, produces a low signal that may easily be discriminated from a
sensor at operating temperature.
The release of hydrogen and availability of oxygen from the secondary air
pump will have two effects on the oxygen sensor depending upon its
temperature. Both effects may be used to establish the operation of the
CHC system and used within the feedback to control the hydrogen release.
If the oxygen sensor is at operating temperature, indicated by a signal
above a threshold value, and hydrogen is released into the exhaust stream,
the signal from the sensor will be depressed due to the presence of
hydrogen. The extent of the signal depression during each release pulse
will correspond to the quantity of hydrogen released from the CHC system
and detected by the oxygen sensor. Self heating oxygen sensors, wherein a
resistive heater is placed in the sensor to bring it to operating
temperature sooner, could be utilized for closed loop control before the
engine has started. The sensor may also be hot when the vehicle has been
operated recently enough that the sensor is still hot, providing an
opportunity to test the CHC system even when the preheating may not be
necessary. Alternatively, this test may be performed after the engine has
reached operating temperature and the release of hydrogen is only to
verify the operation of the CHC system. In another mode of operation the
sensor closest to the engine is monitored during the period of the CHC
preheat where the engine is running and the sensor is just coming to
operating temperature and providing a signal. During this time the sensor
will respond in a sinusoidal or sem-square wave manner as each cylinder
fires and the exhaust is swept past the sensor. The release of hydrogen
could be timed to the firing of the engine so the difference in signal may
be maximized. Additionally, timing the release of hydrogen to the release
of exhaust gas may further improve the utilization of the hydrogen in the
converter, e.g., releasing the hydrogen between exhaust strokes when the
most oxygen is present or utilizing the "back and forth" motion of the
exhaust gas in the system while the engine is running.
In a completely different mode of operation, a cold sensor may be rapidly
brought to operating temperature due to the presence of hydrogen and
oxygen in the exhaust stream. The face of the sensor generally exposed to
the exhaust consists of a porous platinum catalyst that will serve as an
oxygen-hydrogen combination catalyst. The design of the sensor and the
placement of the sensor with respect to the hydrogen inlet(s) and oxygen
inlet(s) may be further optimized to provide the best signal and to
eliminate any damaging or aging affects due to this mode of heating.
During the operation of the CHC system before the cranking of the engine
the initially cold sensor will be monitored. As it heats (due to
hydrogen-oxygen combination) its signal will correspond to the temperature
of the sensor face, and therefore to the amount of hydrogen in the exhaust
stream. When the sensor is fully heated its signal will become
representative of the oxygen content in the exhaust. As the hydrogen
release is varied or pulsed and the oxygen stream is fixed, the sensor
will provide a real time signal as to the amount of hydrogen, in part
being depressed due to the presence of hydrogen. A sensor located after
the first brick (or further downstream) in the converter would respond not
the presence of hydrogen (since it will have reacted on the brick) but the
lowering of the oxygen concentrations resulting from combination with
hydrogen. Following this response will allow the conversion efficiency,
albeit only directly for hydrogen but possibly correlated to other gases,
of the catalytic converter to be monitored. This downstream sensor will
provide a real time response to a release of hydrogen since that release
and combination with the secondary air will cause a significant dip in the
oxygen content measured by the downstream sensor.
This response from the sensor can be characterized and used to establish
the performance of the CHC system (OBD II) as well as used for closed loop
feedback during hydrogen release.
Furthermore, on vehicles equipped with sensors before and after the
converter, the relative response of these two sensors during preheating
and possibly in a special "performance evaluation mode" (when both sensors
are at operating temperature) where H.sub.2 is released into the exhaust
stream during steady state operation of the engine and the calculated
difference in hydrogen and oxygen content in the exhaust is compared to
the measured values.
The use of the oxygen and lambda sensors may require additional hydrogen
injection points so only a partial flow of H.sub.2 is seen by the O.sub.2
sensor (to prevent the sensor from overheating for example).
In a similar manner, a catalyst coated temperature sensor could be used as
a direct measurement of the CHC performance. This sensor would serve as a
representative sample of the catalytic converter, providing a temperature
response as its surface combines the H.sub.2 and O.sub.2. It would be more
effective at evaluating the actual catalyst temperature on the converter
than a temperature sensor simply placed in the air stream responding to
hot air.
Another feedback device would be an optical sensor placed so it is exposed
to the front surface of the main catalyst. This sensor would respond to
the colors of the catalyst face as the system is preheated.
This optical sensor could be used for real time feedback during hydrogen
release to minimize the preheating time by maximizing the hydrogen
release. The controller would release H.sub.2 and slow or stop the release
if a portion of the catalyst were to approach the hydrogen light off
temperature. H.sub.2 release would be controlled to maintain catalyst face
temperature as close to but below this light-off temperature. The sensor
could be provided with an optical filter that only allows colors near the
light off temperature to be detected by the sensor. This would simplify
the electronics since all colors below the threshold temperature would be
ignored automatically. The sensor could be integrating, i.e., respond to
the entire face and produce one signal, or segmented, i.e., provide
signals for each area of catalyst. A segmented sensor could be used with
multiple injection points to actively modify the hydrogen/oxygen
distribution within the converter.
During hydrogen generation in the electrolyzer 22, water is transferred
from the anode side of the electrolyzer 22 over to the cathode side due to
electroosmosis through the proton exchange membrane. This water eventually
collects in the bottom of storage vessel 24 and is withdrawn and/or
recovered using one or two valves 38, 39, e.g., solenoid valves. If two
valves 38, 39 are used, then they may be plumbed in series and have a
captive gas chamber between them. The valves 38, 39 may be opened one at a
time with a fixed amount of water being pushed from the reservoir 24 and
compressing the headspace in the stub during the first half of the cycle
and transferred to the anode water reservoir during the second half of the
cycle. This sequence may be continued until only a small amount of water
remains in the reservoir 24. The water level is sensed using a sensor 146,
e.g., an electro-optic liquid level sensor. Performance verification of
this cathode water recovery system is concerned with the complete opening
and closing of the valves 38, 39 and of the accurate detection of the
water level.
The rate of hydrogen pressure increase during hydrogen generation is
dependent upon the hydrogen storage vessel 24 size. Failure of the cathode
water recovery system to remove water from the hydrogen storage vessel 24
will result in a smaller effective volume for storing hydrogen and a
faster rate of rise of hydrogen gas pressure during electrolysis. A
hydrogen pressure drop of known magnitude during water recovery indicates
that the valves 38, 39 are cycling correctly. Additionally, the quantity
of water recovered must fall within a range that is directly proportional
to the quantity of hydrogen generated.
If a valve 38 or 39, or in some cases both valves 38, 39, fail to open,
there is no effect on safety, but the hydrogen storage capacity is
reduced. This failure may be detected when the hydrogen storage vessel
pressure fails to decrease during an attempted water recovery cycle.
Additionally, an abnormally high pressure increase during hydrogen
generation may indicate this type of failure. The most obvious result is
the water level not falling.
If one of the valves 38 or 39 fails to close, there may be a release of
hydrogen into the anode water reservoir 20. In this condition, the system
10 may continue under limited operation until maintenance can be
performed. A sudden pressure drop as hydrogen gas escapes during a water
recovery step may indicate this type of failure. The ambient hydrogen
detector 337 may also detect the hydrogen that has escaped.
If the valves 38, 39 leak, then there may be a release of hydrogen into the
anode water reservoir, and the system 10 will have to be shut down. A
pressure loss during idle mode may indicate this type of failure, and so
may the ambient hydrogen detector 337.
If the water level detector 146 fails such that it indicates the vessel 24
to be dry, then storage vessel 24 will just continue to store water, thus
reducing the capacity for hydrogen. This failure condition may be
indicated by the recovered cathode water volume being outside the
appropriate range. Also, the rate of increase of the hydrogen pressure may
be abnormal, since the available volume of the hydrogen storage reservoir
is reduced.
If the water level detector 146 fails indicating that the vessel 24 is wet,
then there may be a release of hydrogen into the anode water reservoir.
The system 10 may continue under limited operation until maintenance can
be performed. A bubble detector (not shown) in the recovered water path
may detect this failure. The ambient hydrogen detector 337 may also detect
this failure if hydrogen gas is allowed into the vented anode water tank.
As the eventual source of the hydrogen, it is preferred the quantity and
quality of the anode water must be monitored. Water level and quality
monitoring is provided by conductivity sensors 317, perhaps having dual
conductivity probes mounted at the middle and bottom of the anode
reservoir. The temperature of the electrolyzer 22 is a function of the
ambient temperature, water quality, water quantity and electrolyzer
efficiency. A compromised water delivery system will result in an elevated
operating temperature. Water quality is continuously monitored using the
conductivity probes. When both conductivity probes are immersed, a direct
comparison between the probes provides linearity and relative accuracy
measurements. The conductivity sensors also provide a means of measuring
the water level.
When all of the anode water is consumed or lost, electrolyzer 22 will be
unable to generate hydrogen. High cell voltage measurements and an
eventual drop in stack current may indicate this failure mode. A reduced
number of gas bubbles detected by the bubble detector 37, slow pressure
rise in the storage vessel and a high electrolyzer temperature may also
indicate this type of failure.
When anode water flow is interrupted, the electrolyzer 22 may overheat
requiring shut-down of the system 10. A high electrolyzer 22 temperature
may indicate this failure mode. A reduced rate of gas bubbles occurring at
the bubble detector 37 and a temperature dependent cell voltage
measurement may also indicate this type of failure.
When the conductivity sensors 317 fail, whether they indicate the reservoir
to be wet or dry, the system 10 may continue operation, but with a panel
indicator if necessary. In either case, the conductivity must be within
certain predetermined limits. Increased cell voltage measurements and
increased temperature are indicative of a dry reservoir or a wet reservoir
having a high conductivity fluid.
If the conductivity sensors 317 fail or drift, then possible damage may
occur to electrolyzer 22 due to poor water quality. There may also be a
false panel indicator concerning water quality. Again, the conductivity
must be within certain limits. Increased cell voltage measurements and
increased temperature may be used to verify or identify poor water
quality.
A source of oxygen, in air, for combination with the electrochemically
generated hydrogen to preheat the catalytic converter may be provided by
means of the electrically activated air pump 18. Performance verification
of the hydrogen delivery system 10 preferably includes quantifying the
amount of air delivered by the pump. The air pump can be well
characterized by providing the air flow as a function of current draw.
Measurement of the current supplied to air pump 18 provides the status of
the electronics supporting air pump 18 and the quantity of air being
delivered to the vehicle's exhaust system. This current may be measured
using a shunt or voltage drop across the solid state switch. If there is
no air supplied to the exhaust system, then the hydrogen will have nothing
with which to combine to preheat the catalytic converter, so the system 10
will have to be shut down.
If there is a blown fuse or other electrical failure of air pump 18 or
motor thermal overload of air pump 18, then shut-down of the system 10 may
be required. A lack of any current to air pump 18 may indicate this type
of failure.
A reduced air flow due to a restriction or back pressure in air pump 18 may
mean that there is not enough oxygen to combine with the hydrogen to
preheat the catalytic converter 13, so excess hydrogen may be released
through the vehicle's exhaust system. This condition may result in reduced
heating of the catalytic converter 13. Current draw in air pump 18 may
indicate this failure. The ambient hydrogen detector 337 described earlier
will not help in this case because it is not positioned to detect hydrogen
coming out of the exhaust.
If there is a partial failure of the motor or wiring harness or mechanical
failure in air pump 18, then a shut-down of the system 10 will be
required. Current draw in air pump 18 may indicate this failure. Other,
more conventional means of detecting air flow, e.g., hot wire, vane
switches, etc., could also be utilized.
In the present embodiment, there are several flags that indicate problems,
or errors, of varying degrees of seriousness within the hydrogen delivery
system 10. Various procedures, described below, set these flags and/or
take appropriate action, such as flashing an indicator light for the
driver, shutting the hydrogen delivery system down, or other appropriate
response. When an error is detected, the controller may alert the driver
immediately, or may wait to see if the flag goes away after a period of
time or following a particular event. For example, the controller may wait
for a few driving cycles to see if the flag goes away. A driving cycle
includes starting the vehicle, bringing the vehicle up to normal operating
conditions, and stopping the vehicle. For another flag, the controller may
cause itself to re-initialize and then check to see if the flag is still
present. The programmed response depends on the seriousness of the
indicated problem. The flags may be kept in any convenient storage, such
as an EEPROM, special registers or the memory used by the controller. The
error, or total operating history, of the CHC system may be provided using
fault codes and interfaces similar to existing SAE standards.
The operation of a controller system and on-board-diagnostics is illustrate
in FIG. 17 which shows a general operating flow for the main procedure for
a controller system and on-board-diagnostics. There may be concurrently
running procedures. In other words, controller 134 may be multithreading,
or there may even be another controller. Concurrent threads may be
required to monitor the system 10 for any errors. Therefore, although the
below described procedures may not show a means to monitor for errors
while a procedure is performing a task that could be hampered or made
dangerous by an error, the skilled person will understand that concurrent
processes are readily apparent that could maintain the system 10 at all
times. 5 The first step is to initialize the controller 134 in step 400 to
bring the system up to a known state, wherein all controls and flags are
initially turned off. This step may be done upon powering up the system
when the vehicle's battery is attached to the vehicle, or by a service
technician, or after a certain number of driving cycles, or in response to
an error.
Control then proceeds to step 402 to wait for a key trigger or key-on
signal. This step is also 10 known as the idle state. In this state, the
controller 134 is waiting for a signal over input signal line 314 that the
engine is about to start, which means that hydrogen delivery may soon be
necessary. In the meantime, while the controller 134 is waiting for a
key-on signal, it may be monitoring certain system parameters. This step
will be described in more detail below with reference to FIG. 18.
Control then proceeds to step 404 wherein the controller 134 checks for a
signal indicating 15 that hydrogen is actually to be released. This step
will also do the hydrogen release if it is called for. This step will be
described in more detail below with reference to FIG. 19.
Control then proceeds to step 406 wherein the controller 134 checks to see
if the engine is actually cranking over before proceeding further. This
step will be described in more detail below with reference to FIG. 20. 20
Control then proceeds to step 408 wherein the controller 134 checks for
another signal to actually release hydrogen. This step will also do the
hydrogen release if it is called for. This step will be described in more
detail below with reference to FIG. 21. Control then proceeds to step 410
wherein the controller 134 will cause the electrolyzer 22 to produce
hydrogen if it is needed to refill the hydrogen vessel 24 after any
hydrogen releases in 25 steps 404 and 408. This step will be described in
more detail below with reference to FIG. 22. Control then proceeds to step
412, wherein the controller 134 will check to see if the key-on signal is
still asserted or if any errors occurred in any of the previous steps.
This step will be described in more detail below with reference to FIG.
23.
FIG. 18 describes the process that the controller 134 goes through while it
is waiting for a key-trigger or key-on signal from the ignition during
step 402 of the main procedure. The key trigger is any kind of a
pre-indication signal that the driver is likely to try to start the engine
soon. For example, it could be a signal that the driver has turned the key
in the ignition to the first position, or accessory position. It could be
a signal that the driver has turned the key all the way to crank the
engine with a delay built in before the engine actually starts to crank.
The controller 134 enters the process at step 414, Begin. Control proceeds
to step 416, wherein the controller 134 reads the current tank pressure
(Pc) from the hydrogen pressure sensor 144 through sensor input line 310.
Control then proceeds to step 418, wherein the controller 134 receives data
from serial port 341. This port can be any type of I/O device, not just a
serial port. If controller 134 is part of the vehicle's built-in engine
management unit, then that system will have its own serial port which can
be used. The I/O device may be whatever link is required for communication
with off-board diagnostic equipment. The off-board diagnostic equipment is
a machine or computer that a service technician may use to diagnose
problems with vehicles brought in for service. If the vehicle is not
attached to an off-board diagnostic device, then the controller 134 will
receive nothing from serial port 341 in step 418 and the procedure will
just continue on past this step. An alternative embodiment could have the
procedure on FIG. 18 check to see if the serial port is attached to the
offboard diagnostic device before attempting to read data from it. The
data that may be read from the serial port may be any information that the
off-board diagnostic device or service technician needs to send to the
vehicle. For example, the service technician could adjust operating
parameters or threshold values that change the performance of the vehicle.
In addition to receiving data over the serial port 341, the controller may
send data to the off-board diagnostic device. For example, all of the same
sensor signals that are described herein could be sent to the off-board
diagnostic device for a thorough diagnosis of the error within the
vehicle, so that the service technician may make repairs to the vehicle,
specifically system 10. Additionally, all of the flag data could be sent
to the off-board diagnostic device. The driver of the vehicle may not be
able to repair whatever error occurs within the vehicle, so many errors or
flags will generate nothing more than a generic warning to the driver that
the vehicle needs to be serviced. The service technician, on the other
hand, would need very detailed error information, including a mileage log
before and after the error or flag, so a considerable amount of data sent
may be sent over serial port 341. The flags and sensor data are used in
the diagnostic procedures described herein.
Control then proceeds to step 420, wherein the controller 134 stores any
data from the serial port 341 in memory. This memory may be an EEPROM,
although it is recognized that there are many types of memory devices that
could provide data storage.
Control then proceeds to step 422, wherein the controller 134 checks for
any hydrogen in the atmosphere, which could indicate a leak in the system
10. This step will be described in more detail below with reference to
FIG. 24.
Control then proceeds to step 424, wherein the controller 134 checks to see
if the current tank pressure (Pc) is less than 50 psi. The data for the
tank pressure (Pc) is read from hydrogen pressure sensor 144. If Pc<50
psi, then the controller 134 will set flag (Ee) in step 426.
After the flag (Ee) is set or if Pc is not less than 50 psi, then control
proceeds to step 428, wherein the controller 134 checks to see if the tank
pressure (Pc) is less than the tank setpoint pressure (Psp). If the answer
is yes, then the controller 134 sets the hydrogen leak flag (Ed) in step
430. The tank setpoint pressure is the pressure to which the storage
vessel 24 should be pressurized, and the system should not lose this
pressure while idle.
After the flag (Ed) is set or if Pc is not less than Psp, then control
proceeds to step 432, wherein the controller checks to see if the key-on,
or key trigger, signal is asserted. If not, then control returns to step
416 to begin the above procedure again. Otherwise, control proceeds to
step 434.
In step 434, the controller 134 checks to see if the engine temperature
(Te) is greater than the engine threshold temperature for hydrogen release
(Teth). If Te>Teth, then the vehicle is deemed to have been operated
recently so the catalyst will still be at the light-off temperature and
does not need hydrogen to be released to heat up the catalyst, so the
procedure returns to step 416 to continue waiting and monitoring.
If Te is not greater than Teth, then control proceeds to step 436, wherein
the controller 134 sends a signal on output control line 316 to air pump
18 to direct air pump 18 to turn on in anticipation of a hydrogen release
in one of the next steps. Control then proceeds to step 438, wherein the
procedure returns to the main procedure in FIG. 17 and proceeds to step
404.
FIG. 19 describes the process that the controller 134 takes the system 10
through in step 404 to determine if hydrogen is to be released in response
to a key trigger release or a post key trigger release. The trigger
releases may be software generated signals depending on how many release
steps the controller is programmed to perform. FIGS. 19 and 21 show a
total of four release steps; however, more or fewer steps may actually be
programmed into the controller. These figures are merely illustrative of
an embodiment of the invention. It may even be possible to change the
number of release steps that the controller is programmed to perform, such
as through the data received over serial port 341.
The controller 134 enters the process at step 440, begin. Control proceeds
to step 442, wherein the controller 134 checks to see if a key trigger
release signal has been received.
If there has not been a key trigger release signal, then control proceeds
to step 446. Otherwise, control proceeds to step 444, wherein controller
134 orders hydrogen to be released. The hydrogen release step will be
described below in more detail with reference to FIG. 25. Control then
proceeds to step 446.
In step 446, controller 134 checks to see if a post key trigger release
signal has been received. If there has not been a post key trigger release
signal, then control proceeds to step 450. Otherwise, control proceeds to
step 448, wherein controller 134 orders hydrogen to be released. As stated
above, the hydrogen release step will be described below in more detail
with reference to FIG. 25. Control then proceeds to step 450 wherein the
procedure returns to the main procedure in FIG. 17 and proceeds to step
406.
FIG. 20 describes the process that controller 134 goes through in step 406
while it waits for a crank trigger signal. Crank trigger is the signal
that the driver has turned the key all the way to actually start cranking
the engine. Controller 134 enters the process at step 452, begin. Control
proceeds to step 454, wherein controller 134 checks to see if the elapsed
time since the air pump was started is greater than the air pump idle time
(Tpi), e.g., 15 seconds. If so, then controller 134 sends a signal on
output control line 316 to air pump 18 to direct air pump 18 to turn off
since it is better not to leave the air pump on if the crank trigger
signal is taking a long time to arrive.
After the air pump is directed to stop, or if elapsed time is not greater
than Tpi, then control proceeds to step 458, wherein controller 134 checks
to see if the crank trigger signal has been received. If not, then control
returns to step 454 to continue this process while waiting. Otherwise,
control proceeds to step 460. An alternative embodiment may have another
step if the answer in step 458 was "no", wherein the procedure checks to
see if the key-on signal is still asserted. If not, then the driver may
have aborted the attempt to start the engine, so the procedure should jump
back out to the main procedure. The system may be programmed to perform a
limited number of pre-crank hydrogen releases if the driver does not
actually start the vehicle. For example, the system would allow the driver
to turn the key on and off again three times, releasing hydrogen each time
the converter is cold. The system would then not release hydrogen on the
fourth attempt. This "false start" counter could be reset by a successful
start or timeout period.
In step 460, controller 134 sends a signal on output control line 316 to
air pump 18 to direct the air pump 18 to turn on. If the air pump 18 was
never turned off because the process never reached step 456, then this
signal will have no effect on the air pump 18. Control then proceeds to
step 462, wherein the procedure returns to the main procedure and
continues at step 408.
FIG. 21 describes the process that the system 10 goes through in step 408
to determine if hydrogen is to be released in response to a crank trigger
release or a post crank trigger release. The controller 134 enters the
process at step 464, begin. Control proceeds to step 466, wherein the
controller 134 checks to see if a crank trigger release signal has been
received.
If there has not been a crank trigger release signal, then control proceeds
to step 470. Otherwise, control proceeds to step 468, wherein controller
134 orders hydrogen to be released. The hydrogen release step will be
described below in more detail with reference to FIG. 25. Control then
proceeds to step 470.
In step 470, controller 134 checks to see if a post crank trigger release
signal has been received. If there has not been a post crank trigger
release signal, then control proceeds to step 474. Otherwise, control
proceeds to step 472, wherein controller 134 orders hydrogen to be
released. As stated above, the hydrogen release step will be described
below in more detail with reference to FIG. 25. Control then proceeds to
step 474 wherein the procedure returns to the main procedure in FIG. 17
and proceeds to step 410.
FIGS. 18-21 showed two possible triggers for indicating that hydrogen may
soon be required to be released: (1) at key-on, and (2) at crank of the
engine. It is possible to conceive of other ways to signal an impending
need for hydrogen. For example, the vehicle could even be equipped with a
separate trigger button that the driver could manually operate to signal
that the vehicle will soon be started.
FIG. 22 describes the process by which the system 10 produces hydrogen
through electrolysis in step 410. Controller 134 enters the process at
step 476, begin. Control proceeds to step 478, wherein the controller goes
through a series of preliminary checks to determine certain parameters of
the system 10 and the vehicle. These preliminary checks will be described
in more detail below with reference to FIG. 26.
Control proceeds to step 480, wherein controller 134 performs a routine to
maintain electrical power to the stack of cells 26 within electrolyzer 22.
This routine will be described in more detail below with reference to FIG.
27. The first time the procedure goes through this step, it turns on the
electrolyzer by sending a signal to start hydrogen generation. In an
alternative embodiment, this procedure may also start a timer to time how
long it takes to generate hydrogen. An unusually long time period may
indicate a problem with the hydrogen generation system.
Control then proceeds to step 482, wherein controller 134 monitors certain
parameters in the production of hydrogen. Monitoring of the parameters
will be described in more detail below with reference to FIG. 28.
Control then proceeds to step 484, wherein controller 134 reads the current
tank pressure (Pc) data over sensor input line 310 from hydrogen pressure
sensor 144. Controller 134 compares current tank pressure with the tank
setpoint pressure (Psp). If Pc is not greater than or equal to Psp, then
control returns to step 480 to continue maintaining the power to the
electrolyzer 22 and monitoring the production of hydrogen. If
Pc.gtoreq.Psp, then control leaves this loop and proceeds to step 486.
When Pc.gtoreq.Psp the storage tank is full.
In step 486, controller 134 checks water level sensor 146 over sensor input
line 308 to determine if there is water present in the hydrogen tank, or
storage vessel 24. If the water level is too high, then control proceeds
to step 488, wherein controller 134 sends a command sequence over control
output lines 306, 307 to the water outlet valves 38, 39 to cycle and let
some water out of storage vessel 24. The valves 38, 39 are cycled open and
closed alternately in order to draw out a packet of water of known volume,
so hydrogen will not be accidentally let out, too. This step may be
repeated until the water level sensor 146 no longer senses a water level
that is too high. Then control proceeds to step 490 wherein controller 134
checks the current tank pressure (Pc) again and compares it with the tank
setpoint pressure (Psp). If Pc is still not greater than or equal to Psp,
which can happen whenever water is recovered in step 488, then control
returns to the hydrogen production loop starting at the maintenance of
electrical power to the stack in step 480. Otherwise, if Pc is still
greater than or equal to Psp, then control proceeds to step 492, because
at that point the hydrogen pressure sensor 144 has provided the signal to
the controller 134 that the storage vessel 24 is full.
If when completing step 486 water is not found in storage vessel 24, and
the total water transferred is found within an acceptable range, then
control proceeds to step 492.
When control reaches step 492, this condition indicates that the pressure
of the hydrogen in storage vessel 24 is at a maximum, i.e., the tank is
full, and there is no excess water in storage vessel 24. Therefore,
controller 134 can turn off electrolyzer 22 in this step by sending a stop
hydrogen generation signal over control output line 304 to the
electrolyzer source 138 to turn off. Control then proceeds to step 494,
wherein the process returns control to the main procedure at step 412.
FIG. 23 describes the test complete procedure in step 412 of the main
procedure. Control enters this procedure at step 496, begin. Control then
proceeds to step 498, wherein controller 134 checks whether the key-on
signal is asserted. If so, then this signal indicates that the driver of
the vehicle is either still trying to start the car or the car is actually
running, so the procedure must continue to step 500, wherein control
returns to the main procedure at step 402 in order to proceed with the
above described routines.
If the key-on signal is not asserted in step 498, then control proceeds to
step 502, wherein controller 134 reads the variables from the various
sensors and flags.
Control then proceeds to step 504, wherein controller 134 checks whether
any error flags are set. These flags may have been set by any of the other
steps in the main procedure, or by an independently running thread that is
monitoring system conditions. In either case, the indicated errors may not
have required immediate shut-down of the system 10. If no error flag is
set, then controller 134 continues looping back through step 498. If an
error flag is set, then control proceeds to step 506 wherein controller
134 sends a signal to the panel indicator light to flash on and warn the
driver of the vehicle that an error has occurred and the vehicle's
hydrogen injection system may need servicing. The warning, however, may
just be a low priority service warning. After the procedure has given this
warning, control returns to step 498 to continue the loop. Thus, the
system continues to monitor itself while the vehicle is turned off. The
exact timing of when the panel indicator is activated and the mode of
indication are variables that can be changed as desired.
FIG. 24 shows the procedure that hydrogen delivery system 10 goes through
in a check for a hydrogen leak from the system. This procedure is shown to
be called by another procedure and to return to the other procedure;
however, in an alternative embodiment, this procedure continues to run
independently of any other procedures as an independent thread in the
controller 134. Control is shown to enter this procedure at step 508,
begin. Control proceeds to step 510, wherein controller 134 checks the
signal from ambient hydrogen detector 337 to determine whether hydrogen is
present in the atmosphere. If hydrogen is not detected in the atmosphere,
then control proceeds to step 512, wherein the procedure returns control
to whichever procedure called it.
If hydrogen is detected in step 510, then control proceeds to step 514,
wherein controller 134 sets the hydrogen leak flag (Ef). Control then
proceeds to step 516, wherein controller 134 calls the routine to shut
down the hydrogen delivery system. The shut-down routine will be described
in more detail below with reference to FIG. 29. Although this shut-down
procedure shows the procedure returning to its calling procedure, there
may be certain types of errors/flags that require a complete and immediate
shut-down of the system 10, including the controller 134, so that no
procedure continues running.
FIG. 25 shows the procedure that the hydrogen delivery system 10 goes
through to release hydrogen to the vehicle's exhaust system. Control
enters the procedure at step 518, begin. Control then proceeds to step
520, wherein controller 134 waits for a hydrogen release delay (trd) time
period. Control then proceeds to step 522, wherein controller 134
determines which mode, either timed release mode or volume release mode,
is requested. If a timed release mode is requested, then control proceeds
to step 524. If a volume release mode is requested, then control proceeds
to step 526. The mode may be programmed into the controller 134 such that
controller 134 requests a mode based upon where this procedure was called.
For example, controller 134 may be programmed to request one mode for the
key trigger release step 442, and it may be programmed to request another
mode for the post key trigger release step 446, or the crank trigger
release step 466, or the post crank trigger release step 470. The choice
of modes may even be reprogrammed through the data received from serial
port 341, self tuning, or based on other parameters (such as ambient
temperature, driving habits and preferences, etc.). Additional release
modes, such as closed loop modes based on sensor data such as exhaust
oxygen sensors, temperature sensors, optical temperature sensors, etc.,
are not fully described but may be easily envisioned and implemented.
In step 524, controller 134 determines which type of timed release method
is requested; continuous release, packet release or pulsed release. As
with the choice of modes, the choice of release methods is also programmed
into controller 134 depending on where the hydrogen release procedure was
called from.
For a timed continuous release, control proceeds to step 528, wherein both
valves 46, 48 are held open for a certain amount of hydrogen release time
(tvr). Afterwards, control proceeds to step 534. For a timed packet
release, control proceeds to step 530, wherein packets of hydrogen are
released over a period of time. This timed packet release method will be
described in greater detail below with reference to FIG. 30. Afterwards,
control proceeds to step 534. For a timed pulsed release, control proceeds
to step 532, wherein pulses of hydrogen are released over a certain period
of time. This timed pulsed release method will be described in greater
detail below with reference to FIG. 31. Afterwards, control proceeds to
step 534.
In step 526, controller 134 determines which type of volume release method
is requested; continuous release, packet release or pulsed release. For a
volume continuous release, control proceeds to step 536, wherein both
valves 46, 48 are held open until a certain volume of hydrogen is release
(Vr). This volume may be calculated by controller 134 based on the change
in pressure in the storage vessel 24 as measured by hydrogen pressure
sensor 144. Afterwards, control proceeds to step 534. For a volume packet
release, control proceeds to step 538, wherein packets of hydrogen are
released until a certain volume of hydrogen is released. This volume
packet release method will be described in greater detail below with
reference to FIG. 32. Afterwards, control proceeds to step 534. For a
volume pulsed release, control proceeds to step 540, wherein pulses of
hydrogen are released until a certain volume of hydrogen is released. This
volume pulsed release method will be described in greater detail below
with reference to FIG. 33. Afterwards, control proceeds to step 534.
In step 534, controller 134 reads the current tank pressure (Pc) and
compares it to the tank pressure at the beginning of the hydrogen release
(Pbr). If Pc is not less than Pbr, then controller 134 sets an error flag
(E7) and control proceeds to step 544. If Pc is less than Pbr, then
control proceeds to step 544. In step 544, controller 134 returns control
to whichever procedure called for this hydrogen release procedure.
FIG. 26 shows the preliminary checks procedure that hydrogen delivery
system 10 goes through to check certain parameters of the system prior to
producing hydrogen. Control enters the procedure at step 546, begin.
Control proceeds to step 547, wherein a timer is started. Control proceeds
to step 548, wherein controller 134 checks to see if the time elapsed
(time) since starting the timer in step 547 exceeds a certain time period
after which the air pump should be turned off (tpoff). If so, then control
proceeds to step 550, wherein controller 134 can go ahead and turn off air
pump 18 by sending a signal over output control line 316, and then control
proceeds to step 552. If "time" is not greater than tpoff, then control
returns back to step 548.
In step 552, controller 134 waits for a hydrogen release delay time (trd).
Then control proceeds to step 554, wherein controller 134 checks to see if
the temperature of the engine (Te) is greater than the engine threshold
temperature for electrolyzing (Tethe). If Te is not greater than Tethe,
then control loops back to this step to wait for the temperature to come
up high enough for electrolyzer to begin production of hydrogen. If
Te>Tethe, then control proceeds to step 556. In an alternative embodiment,
if Te.ltoreq.Tethe, then there may be an additional step wherein the
procedure checks to see if the key turns off, so the procedure can jump
out and go to idle.
In step 556, the controller 134 checks to see if the system volts (Es) of
the vehicle's electrical system is greater than the minimum system volts
(Esmin) required for operation of the electrolyzer. If not, then
controller 134 waits for a battery recovery time (tbr) in step 558 and
then checks to see if the system volts (Es) is still not greater than the
minimum system volts (Esmin) in step 560. If Es is still not greater than
Esmin, then controller 134 sets an error flag (E3) in step 562 and then
returns to the idle state in step 564. The independently running procedure
mentioned above for monitoring errors and flags may be able to take the
appropriate action for this error flag, since the preliminary checks
procedure is returning to idle. If Es was greater than Esmin in either
step 556 or step 560, then control proceeds to step 566. In an alternative
embodiment, if Es.ltoreq.Esmin in step 560, then the procedure could loop
back to step 558 and continue to wait for the system volts to come up. In
the same loop, the procedure could monitor the key-on signal to see if the
vehicle is ever turned off, so the procedure could go to idle.
In step 566, controller 134 checks to see if the resistivity of the water
is less than 1 M ohm.cm, indicating that the water may be contaminated. If
so, then controller 134 sets an error flag (E6) in step 568, and then
control proceeds to step 570. If not, then control proceeds directly to
step 570.
In step 570, controller 134 checks to see if there is water in the anode
reservoir. If there is no water, then controller 134 sets an error flag
(E8) in step 572, and control proceeds to step 574. If there is water,
then controller 134 proceeds directly to step 574, wherein the procedure
returns control to the procedure that called it, such as the electrolyze
procedure in FIG. 22 at step 478. FIG. 27 shows the routine, or procedure,
that hydrogen delivery system 10 goes through to maintain the power to the
stack of cells 26 within electrolyzer 22. Control enters the procedure at
step 576, begin, and proceeds to step 578, wherein controller 134 sets the
current supplied to the stack of cells 26 by adjusting the duty cycle of
the pulse width modulator (PWM). This procedure to set the stack current
will be described in greater detail below with reference to FIG. 34.
Control then proceeds to step 580, wherein controller 134 checks to see if
the individual cell voltages (Vc) are within a specified range, such as
between about 1.5 volts and about 2.5 volts. If not, then controller 134
sets an error flag (E9) in step 582, and control proceeds to step 584.
In step 584, controller 134 checks to see if the duty cycle of the PWM is
less than 50%, indicating that the stack current required some adjustment
to get it to the proper level. If so, then controller 134 sets an error
flag (Eb) in step 586, and control proceeds to step 588. If the PWM's duty
cycle was not less than 50%, then control proceeds directly to step 588.
In step 588, controller 134 checks to see if the current of the stack
(Istk) is less than 20 Amps and the temperature of the water (Tw) is
greater than or equal to 50.degree. C. If both conditions are true, then
there is an efficiency problem in the stack because the water temperature
should be closer to ambient temperature when the current is that low, so
controller 134 sets an error flag (E4) in step 590, and control proceeds
to step 592. If one condition is false, then control proceeds directly to
step 592, wherein the procedure returns control to the electrolysis
process in FIG. 22 at step 482.
FIG. 28 shows the procedure that hydrogen delivery system 10 goes through
in step 482 of the electrolysis process in FIG. 22 to monitor certain
parameters during the production of hydrogen. Control starts at step 594
with the monitoring of the gas bubble frequency in return line 34 through
sensor input line 312. If the bubble frequency is not within the proper
range, such as between about 20 and about 100 bubbles per minute, then
controller 134 sets an error flag (E5) in step 596, and control proceeds
to step 598. If the bubble frequency is within the proper range, then
control proceeds to step 598.
In step 598, controller 134 checks to see if the temperature of the water
(Tw) is greater than 20.degree. C., and if not, then controller 134 sets
an error flag (El) in step 600 indicating that the water is too cool
considering that the temperature should rise during electrolysis, so
something may not be working right if the temperature is too low. If the
temperature of the water is greater than 20.degree. C., or after flag E1
is set, control proceeds to step 602.
In step 602, controller 134 checks to see if the temperature of the water
(Tw) is greater than 85.degree. C., and if not then control proceeds to
step 612. Otherwise, controller 134 sets an error flag (Ea) in step 604
and proceeds to check whether Tw is even greater than 95.degree. C. in
step 606, and if not, then the water is hot, but not too hot, so control
can proceed on to step 612. Otherwise, the temperature of the water is too
hot, so controller 134 sets error flag (Ea), which may be a different flag
than the one set in step 604, in step 608 and then goes to the shut down
procedure in step 610 since something must be wrong with the system if the
water temperature is so high. This shut down procedure is described in
greater detail below with reference to FIG. 29.
In step 612, controller 134 checks to see if the change in pressure over
time (dP/dt) is greater than zero or within a window. Controller 134 may
save the current pressure and time since the last reading in memory in
order to calculate the change. If the pressure is not increasing, then
controller 134 sets an error flag (Ec) in step 614. Afterwards or
otherwise, the procedure returns in step 616 to step 484 in the
electrolysis process in FIG. 22.
FIG. 29 shows the procedure that the hydrogen delivery system 10 goes
through to shut down the system 10, usually in response to an error in the
system 10 that could cause serious problems if the system 10 continued to
operate. Control enters the procedure at step 618, begin, and then
controller 134 proceeds to send a signal over output control line 316 to
turn off (close) the air pump 18 in step 620, to send a signal over output
control line 304 to turn off the electrolyzer 22 in step 622, and to send
signals over output control lines 306, 300 and 302 to turn off all valves
38, 46 and 48 in step 624. Then in step 626, the procedure returns the
system 10 to the idle state in the main procedure at step 402 in FIG. 17.
FIG. 30 shows the procedure that the hydrogen delivery system 10 goes
through to release hydrogen in timed packets for a specified time period
in step 530 of FIG. 25. Control enters the procedure at step 628, begin,
and controller 134 proceeds to open valve 46 for a time period, e.g. 50
ms, and then closes it in step 630. Then controller 134 waits for another
time period, e.g. another 50 ms, in step 632. This extra wait step gives
the valve a chance to close and become seated before opening valve 48 for
a time period, e.g., 50 ms again, in step 634. Then controller 134 waits
for another time period in step 636, another 50 ms is sufficient for the
valve to become seated, and then checks in step 638 to see if the elapsed
time (time) for the duration of this packet release procedure has reached
the desired hydrogen release time (tvr). If not, then the procedure
continues back at step 630 to release another packet. Otherwise, the
procedure returns in step 640 to the hydrogen release procedure in FIG. 25
at step 534.
FIG. 31 shows the procedure that the hydrogen delivery system 10 goes
through to release hydrogen in timed pulses for a specified time period in
step 532 of FIG. 25. Control enters the procedure at step 642, begin, and
controller 134 opens one of the valves 46, 48, e.g., valve 48, in step 644
and holds it open. Controller 134 then opens the other valve, e.g., valve
46, for valve on time (tvon) and closes it in step 646. Controller 134
waits for valve off time (tvoff) in step 648 before checking in step 650
to see if the elapsed time (time) during this entire timed pulsed release
procedure has reached the desired hydrogen release time (tvr). If not,
then the procedure continues back at step 646 to release pulses of
hydrogen, until the hydrogen release time (tvr) has passed, and controller
134 can close valve 48 in step 652. In step 654, the procedure returns to
the hydrogen release procedure at step 534 in FIG. 25.
FIG. 32 shows the procedure that hydrogen delivery system 10 goes through
to release hydrogen in timed packets for a specified volume of hydrogen in
step 538 of FIG. 25. Control enters the procedure at step 656, begin, and
controller 134 proceeds to open valve 46 for a time period, e.g., 50 ms,
and then closes it in step 658. Then controller 134 waits for another time
period, e.g., another 50 ms to let the valve seat, in step 660 before
opening valve 48 for a time period, e.g., 50 ms again, in step 662. Then
controller 134 waits for another time period in step 664, another 50 ms is
sufficient to let the valve seat, and then checks in step 666 to see if
the volume of hydrogen released for the duration of this packet release
procedure has reached the desired hydrogen release volume (Vr). The volume
of hydrogen released may be calculated by the controller 134 from the
pressure change of the storage vessel and summing the volume of each
packet. If volume released<Vr, then the procedure continues back at step
658 to release another packet. Otherwise, the procedure returns in step
668 to the hydrogen release procedure in FIG. 25 at step 534.
FIG. 33 shows the procedure that the hydrogen delivery system 10 goes
through to release hydrogen in timed pulses for a specified volume of
hydrogen in step 540 of FIG. 25. Control enters the procedure at step 670,
begin, and controller 134 opens one of the valves 46, 48, e.g., valve 48,
in step 672 and holds it open. Controller 134 then opens the other valve,
e.g., valve 46, for valve on time (tvon) and closes it in step 674.
Controller 134 waits for valve off time (tvoff) in step 676 before
checking in step 678 to see if the volume of hydrogen released during this
entire volume pulsed release procedure has reached the desired hydrogen
release volume (Vr). If not, then the procedure continues back at step 674
to release pulses of hydrogen, until the hydrogen release volume (Vr) has
passed, and controller 134 can close valve 48 in step 680. In step 682,
the procedure returns to the hydrogen release procedure at step 534 in
FIG. 25.
FIG. 34 shows the procedure that hydrogen delivery system 10 goes through
to set the current (Istk) delivered to the stack of cells 26. Control
enters the procedure at step 684, begin, and controller 134 proceeds to
set the duty cycle of the pulse width modulator (PWM) to 50% in step 686.
Then controller 134 determines in step 690 whether the stack current
(Istk) is less than, equal to, or greater than the maximum stack current
(Istkma). The system prefers for Istk to be in a range fairly close to
Istkma, so the controller 134 will increment the PWM's duty cycle by about
1% in step 692 if Istk is too low, or decrement the PWM's duty cycle by
about 1% in step 696 if Istk is too high, and either way return to step
690. When Istk finally is close enough to Istkma, the procedure returns
control in step 694 to the stack power maintenance procedure at step 580
in FIG. 27. Istkma may depend on the configuration of the stack, the
condition of the battery or the ambient temperature. If Istk is too high,
it could damage the vehicle's alternator. If Istk is too high or low, then
there may be something wrong with the hydrogen delivery system and/or it
will take too long to restore the hydrogen supply.
FIG. 35 shows the procedure, running as an independent thread, that the
controller 134 may go through to periodically update data to send via the
serial port 341 in case an off-board diagnostic device or computer is
attached to the vehicle. A timer is started in step 698. Then in step 700,
the controller 134 waits for a time period, e.g., 2 sec. Then in step 702,
the controller 134 takes the current data and sends it on the serial port
341 to the off-board diagnostic device. Then the timer is reset in step
704, and the procedure returns to step 698 for a continuous loop through
these steps, so the status of the sensors are captured every two seconds
or so.
While the foregoing is directed to the preferred embodiment of the present
invention, other and further embodiments of the invention may be devised
without departing from the basic scope thereof, and the scope thereof is
determined by the claims which follow.
Top